nhc.gov.cn. Shadow Identity

_________2021_1-6__CSV____hqmsts_M20210219_1_1__.csv

14 columns5,666 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

_________2021_1-6__CSV____hqmsts_M20210316_1_2__.csv

14 columns5,004 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

_________2021_1-6__CSV____hqmsts_M20210427_1_3_.csv

14 columns5,773 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

_________2021_1-6__CSV____hqmsts_M20210521_1_4_.csv

14 columns6,102 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

_________2021_1-6__CSV____hqmsts_M20210615_1_5_.csv

14 columns6,285 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

_________2021_1-6__CSV____hqmsts_M20210713_1_6__.csv

14 columns5,548 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

_________2021_7-9__CSV____hqmsts_M20210814_107.csv

14 columns6,667 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

_________2021_7-9__CSV____hqmsts_M20210910_108.csv

14 columns6,521 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

_________2021_7-9__CSV____hqmsts_M20211018_109.csv

14 columns5,371 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

_________HQMSTS_20201205.CSV

10 columns803 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	medium	[8] header 'a11', garbled Chinese text values consistent with patient name field in HQMS medical records schema
10	dob	high	[10] header 'a13', values are YYYY-MM-DD dates (e.g. 1940-06-13, 1949-12-22) consistent with date of birth
12	country	medium	[12] header 'a15c', garbled values represent Chinese nationality field (中国), consistent with HQMS patient nationality
17	ssn	high	[17] header 'a20', values are 18-character strings (e.g. 46010019400613152X) matching Chinese national ID number format, SSN equivalent
20	address1	medium	[20] header 'a24', garbled Chinese text consistent with street address field in HQMS schema
22	address2	medium	[22] header 'a26', garbled Chinese text consistent with detailed/registered address, secondary address line
23	phone	high	[23] header 'a27', values include '66183830' and '13307590908' matching Chinese phone number patterns
26	phone	high	[26] header 'a30', values identical pattern to a27 (e.g. '66183830', '13307590908') — secondary/emergency contact phone
28	fullName	medium	[28] header 'a32', garbled Chinese text consistent with next-of-kin or emergency contact name in HQMS schema
31	phone	high	[31] header 'a35', values like '13976246333', '17789833393' are clearly Chinese mobile phone numbers

Notes: 291 columns total. File is a Chinese HQMS hospital quality monitoring record with mostly clinical/diagnostic fields. Text is garbled due to encoding issues (likely GB2312/GBK displayed as latin1). Column names follow HQMS internal coding scheme (a01-a49, b-series, c-series). PII identified includes patient name (a11), DOB (a13), nationality (a15c), national ID/SSN (a20), address fields (a24, a26), phone numbers (a27, a30, a35), and contact person name (a32). Gender field a12c contains only 0 values in sample — mapped cautiously as skip. Age field a14 contains integer ages, not DOB. Bulk of columns are ICD-10 diagnosis codes, procedure codes, and clinical metrics.

_________HQMSTS_20201210.CSV

13 columns711 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'a11', short garbled Chinese character strings consistent with patient given names (2-4 chars typical of Chinese personal names)
9	gender	medium	[9] header 'a12c', values are '0' — likely gender code (0=male in Chinese HQMS medical coding)
10	dob	high	[10] header 'a13', values are YYYY-MM-DD dates (e.g. 1972-08-08) consistent with patient dates of birth
12	country	medium	[12] header 'a15c', garbled values decode to '中国' (China) — country of patient
17	ssn	high	[17] header 'a20', 18-digit values (e.g. 460100197208080357) are Chinese Resident Identity Card numbers — national ID equivalent to SSN
18	city	medium	[18] header 'a22', garbled Chinese strings at province/city level (e.g. 广东省), patient registered city/region
20	address1	medium	[20] header 'a24', garbled longer Chinese strings with street/road markers — patient street address line 1
22	address2	medium	[22] header 'a26', garbled shorter Chinese address strings — patient address detail / line 2
23	phone	high	[23] header 'a27', values include Chinese landline (66215202) and mobile numbers (13379822781, 18976250985) — patient phone
26	phone	high	[26] header 'a30', same phone number values as a27 — emergency/contact phone number
28	fullName	medium	[28] header 'a32', short garbled Chinese character strings — emergency contact full names
30	address1	medium	[30] header 'a34', garbled longer Chinese address strings — emergency contact address
31	phone	high	[31] header 'a35', values are Chinese mobile numbers (18976238000, 13337645195) — emergency contact phone

Notes: 291-column Chinese HQMS medical record file. Column names are opaque HQMS internal codes (a01, b12, c06x01c, etc.). Text values are garbled due to character encoding issues (likely GB2312/GBK read as Latin-1). PII fields identified by value pattern analysis: patient name (a11), gender (a12c), DOB (a13), country (a15c), national ID/SSN (a20), city (a22), address (a24, a26), phones (a27, a30, a35), emergency contact name (a32), emergency contact address (a34). Remaining columns are clinical data: ICD-10 diagnosis codes, procedure codes, admission/discharge timestamps, hospital quality metrics, and surgical records — all skipped.

_________HQMSTS_20201215.CSV

9 columns751 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'a11', values are garbled-encoded Chinese patient names (encoding corruption of Chinese characters)
10	dob	high	[10] header 'a13', values are YYYY-MM-DD dates (1945-06-01, 1958-02-07) consistent with patient birth dates
12	country	medium	[12] header 'a15c', repeated garbled values consistent with 中国 (China) — patient nationality field
17	ssn	high	[17] header 'a20', 18-character alphanumeric strings (e.g. 46002119450601641X) matching Chinese National ID (居民身份证) format
18	state	medium	[18] header 'a22', garbled Chinese values consistent with province/region names (上海市, 海南省, 山东省)
20	address1	medium	[20] header 'a24', garbled Chinese values consistent with street/address strings for patient home address
23	phone	high	[23] header 'a27', values include 11-digit Chinese mobile numbers (13976730806) and landline numbers — contact phone field
26	phone	high	[26] header 'a30', values include 11-digit Chinese mobile numbers (13976730806) — secondary contact phone field
31	phone	high	[31] header 'a35', values are 11-digit Chinese mobile numbers (13807665086, 13098989339) — emergency contact or next-of-kin phone

Notes: 291-column Chinese medical records file from HQMS system. Columns use internal coded headers (a11, a13, a20 etc.). Text is heavily mojibake-encoded (UTF-8 Chinese read as Latin-1). Key PII: patient name (a11), DOB (a13), national ID/SSN (a20), province (a22), address (a24), and multiple phone fields (a27, a30, a35). Remaining columns are clinical data: ICD diagnosis codes, procedure codes, ward/department codes, admission/discharge timestamps, and medical quality metrics.

_________HQMSTS_20201220.CSV

14 columns688 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'a11' in HQMS schema is patient name; values are short Chinese character strings consistent with personal names
9	gender	medium	[9] header 'a12c' in HQMS schema is gender code; values are '0' consistent with gender coding
10	dob	high	[10] header 'a13' in HQMS schema is date of birth; values are YYYY-MM-DD dates (e.g. 1947-06-11, 1936-09-11)
12	country	medium	[12] header 'a15c' in HQMS schema is nationality/country; repeated garbled values consistent with '中国' (China)
17	ssn	high	[17] header 'a20' in HQMS schema is Chinese national ID (身份证); 18-digit values like '460100194706111247' are national identity numbers
18	state	medium	[18] header 'a22' in HQMS schema is province/region of household registration; garbled Chinese province name strings
20	address1	medium	[20] header 'a24' in HQMS schema is home address; garbled Chinese strings with street/road markers
21	zip	high	[21] header 'a25c' in HQMS schema is postal code; 6-digit values like '570000', '570100' are Chinese postal codes
22	address2	medium	[22] header 'a26' in HQMS schema is work/unit address; garbled Chinese address strings
23	phone	high	[23] header 'a27' in HQMS schema is patient phone; 11-digit values like '13876023028' are Chinese mobile numbers
26	phone	high	[26] header 'a30' in HQMS schema is contact phone; same 11-digit mobile number pattern
28	fullName	medium	[28] header 'a32' in HQMS schema is emergency contact name; short garbled Chinese name strings
30	address1	medium	[30] header 'a34' in HQMS schema is emergency contact address; garbled Chinese address strings
31	phone	high	[31] header 'a35' in HQMS schema is emergency contact phone; 11-digit Chinese mobile numbers

Notes: 291 columns total from China NHC HQMS medical records. Column headers use HQMS internal codes (a11, a13, a20, etc.). Text values are garbled due to character encoding issues (likely GBK/GB2312 read as Latin-1). Key PII: patient name (a11), DOB (a13), gender (a12c), national ID/SSN (a20), address (a24/a26/a34), postal codes (a25c), phone numbers (a27/a30/a35), emergency contact name (a32) and address (a34). Remaining columns are clinical/diagnostic codes, ICD codes, procedure codes, timestamps, department codes, and hospital quality metrics — all skipped.

_________HQMSTS_20201225.CSV

13 columns926 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'a11', short garbled Chinese strings consistent with patient names in HQMS medical records
10	dob	high	[10] header 'a13', values are ISO date strings like '1949-07-09', '1974-03-24' — clearly dates of birth
12	country	medium	[12] header 'a15c', repeated garbled values consistent with country field (China) in patient records
17	ssn	high	[17] header 'a20', values are 18-character Chinese National ID numbers (e.g. '46003219490709336X') — national identity document equivalent to SSN
18	state	medium	[18] header 'a22', garbled Chinese strings of varying length consistent with province/state of residence
20	address1	medium	[20] header 'a24', longer garbled Chinese strings consistent with home/residential address lines
21	zip	high	[21] header 'a25c', 6-digit values like '570000', '570100' are Chinese postal codes
22	address2	medium	[22] header 'a26', garbled strings consistent with a secondary/current address field alongside a24
23	phone	high	[23] header 'a27', values are 11-digit Chinese mobile numbers (e.g. '13379900819', '13807685768')
24	zip	medium	[24] header 'a28c', 6-digit values like '570000', '460100' — postal codes associated with contact address
26	phone	high	[26] header 'a30', values are 11-digit Chinese mobile numbers matching contact/emergency phone field
31	phone	high	[31] header 'a35', values are 11-digit Chinese mobile numbers (e.g. '13876726467') — likely emergency contact phone
190	fullName	medium	[190] header 'b32', short garbled Chinese strings consistent with medical staff or attending physician names

Notes: 291-column HQMS medical record CSV with heavy mojibake encoding of Chinese characters. Key PII: patient name (a11), DOB (a13), country (a15c), Chinese National ID/SSN (a20), province (a22), home address (a24), postal codes (a25c, a28c), phone numbers (a27, a30, a35), and staff name (b32). Remaining columns are clinical codes, diagnosis fields, procedure codes, timestamps, and internal hospital quality metrics — all skipped.

_________HQMSTS_20201231.CSV

14 columns1,095 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'a11', values are short Chinese character sequences consistent with patient full names
10	dob	high	[10] header 'a13', values are YYYY-MM-DD dates (e.g. 1940-07-12, 1958-08-02) — patient dates of birth
12	country	high	[12] header 'a15c', garbled values decode to 中国 (China) — country of patient
15	gender	high	[15] header 'a19c', values are 1/1/1 — numeric gender code (1=male, 2=female) in HQMS schema
17	ssn	high	[17] header 'a20', 18-digit values (e.g. 460024194007124810) are Chinese Resident Identity Card numbers — national identifier equivalent to SSN
20	address1	high	[20] header 'a24', Chinese text with street numbers (e.g. 338号) — patient home address line 1
21	zip	high	[21] header 'a25c', 6-digit codes (e.g. 571500, 460108) — Chinese postal codes for patient address
22	address2	medium	[22] header 'a26', Chinese address text similar to a24 — secondary/permanent address line
23	phone	high	[23] header 'a27', values like 13876306650 are 11-digit Chinese mobile phone numbers
24	zip	medium	[24] header 'a28c', 6-digit codes (e.g. 570000) — postal code for secondary address
26	phone	high	[26] header 'a30', values like 13078990845 are 11-digit Chinese mobile numbers — alternate/contact phone
28	fullName	high	[28] header 'a32', short 2-3 character Chinese sequences consistent with emergency contact person names
30	address1	medium	[30] header 'a34', Chinese address strings with province/street detail — emergency contact or permanent address
31	phone	high	[31] header 'a35', values like 18976008038 are 11-digit Chinese mobile numbers — emergency contact phone

Notes: 291-column HQMS medical records CSV with cryptic field codes. Most columns are clinical/diagnostic data (ICD codes, procedure codes, timestamps). PII is concentrated in the a1x/a2x/a3x block: patient name (a11), DOB (a13), country (a15c), gender (a19c), national ID/SSN (a20), addresses (a24/a26/a34), postal codes (a25c/a28c/a31c), and phone numbers (a27/a30/a35). Emergency contact name in a32. Values are garbled due to encoding (likely GB2312/GBK displayed as UTF-8).

__________2021-07.csv

10 columns7,323 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
6	ssn	medium	[6] A47: value '410222195710114546' is an 18-digit Chinese national ID number (some rows show '-'), maps as SSN equivalent
8	fullName	medium	[8] A11: short garbled Chinese strings consistent with patient names
9	gender	high	[9] A12C: values '1','2' — standard gender code (1=male, 2=female) in Chinese medical records
10	dob	high	[10] A13: values match YYYY-MM-DD date pattern, consistent with birthdates
16	ssn	high	[16] A20: 18-digit Chinese resident ID numbers (身份证号), the primary national identity number
17	city	medium	[17] A22: garbled Chinese text consistent with province/city-level address fields
19	address1	medium	[19] A24: garbled Chinese text consistent with detailed street-level address
22	phone	high	[22] A27: 11-digit numbers matching Chinese mobile phone format (e.g. 13503780089)
27	fullName	medium	[27] A32: short garbled Chinese strings consistent with contact/next-of-kin names
30	phone	high	[30] A35: 11-digit numbers matching Chinese mobile phone format (e.g. 18637878101)

Notes: 281 columns total; column headers are opaque HQMS field codes. PII identified by value patterns. Chinese text is garbled due to encoding issues. Column 16 (A20) contains 18-digit Chinese national ID numbers which encode DOB and gender but are mapped as ssn. Column 6 (A47) appears to be an alternate/prior ID number. Columns 170-177 (B22-B29) contain short Chinese name-like strings for medical staff roles but are not patient PII and are skipped.

__________2021-08.csv

10 columns4,981 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', short Chinese-encoded text values consistent with patient names in HQMS schema
9	gender	high	[9] header 'A12C', values '1'/'2' are gender codes in HQMS (1=male, 2=female)
10	dob	high	[10] header 'A13', values are YYYY-MM-DD dates matching patient birth dates
12	country	high	[12] header 'A15C', garbled values consistently encode '中国' (China), nationality/country field
16	ssn	high	[16] header 'A20', 18-character Chinese national ID numbers (居民身份证号)
17	address1	high	[17] header 'A22', Chinese-encoded strings representing patient residential address
19	address2	medium	[19] header 'A24', longer Chinese-encoded address strings, likely detailed/mailing address
22	phone	high	[22] header 'A27', 11-digit numbers matching Chinese mobile phone format (13x/15x/18x prefix)
25	phone	high	[25] header 'A30', 11-digit Chinese mobile numbers (emergency/secondary contact phone)
30	phone	high	[30] header 'A35', 11-digit Chinese mobile numbers (additional contact phone)

Notes: 281-column HQMS medical records file with coded column headers. Text values are garbled due to character encoding issues (likely GBK/GB2312 displayed as mojibake). PII includes patient name (A11), gender code (A12C), DOB (A13), nationality (A15C), national ID/SSN (A20), address fields (A22, A24), and multiple phone numbers (A27, A30, A35). Columns B29/A32 contain names but appear to be staff/contact names co-mingled with patient names — excluded due to ambiguity. Remaining columns are clinical/diagnostic codes, ICD codes, procedure codes, timestamps, and hospital quality metrics.

___________TCMMS44037___________2021_01.csv

13 columns1,301 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
4	ssn	medium	[4] JKKH values include 18-digit Chinese ID numbers (e.g. 440301197504292710) which are national ID/SSN equivalents containing DOB and identity info
7	fullName	high	[7] XM is the standard Chinese medical record field for patient name (姓名); values are garbled due to encoding but represent person names
8	gender	high	[8] XB is the standard field for sex/gender (性别) in Chinese medical records; values 1/2 represent male/female
9	dob	high	[9] CSRQ (出生日期 = date of birth); values are YYYY-MM-DD dates consistent with patient birth dates
11	country	high	[11] GJ (国籍 = nationality/country); values are ISO country codes (CHN)
23	ssn	high	[23] ZJHM (证件号码 = ID document number); values are 18-digit Chinese national ID numbers (e.g. 440301197504292710) — primary national identity number
26	address1	high	[26] XZZ (现住址 = current residential address); values are full street addresses (garbled encoding but clearly address data)
27	phone	high	[27] DH (电话 = telephone); values are 11-digit Chinese mobile numbers (e.g. 18923745230)
28	zip	high	[28] YB1 (邮编 = postal code); values are 6-digit Chinese postal codes (e.g. 518001)
29	address1	medium	[29] HKDZ (户口地址 = household registration address); values are full addresses — secondary address field for patient
30	zip	medium	[30] YB2 (邮编2 = second postal code corresponding to HKDZ)
34	fullName	high	[34] LXRXM (联系人姓名 = contact person name); values are person names (next of kin/emergency contact)
37	phone	high	[37] DH1 (电话1 = second telephone number); values are 11-digit Chinese mobile numbers (e.g. 13530000346)

Notes: 269-column Chinese medical records file from HQMS system. Encoding is corrupted (mojibake) for Chinese characters but field semantics are clear from standard HQMS schema. Column 4 (JKKH) and column 23 (ZJHM) both contain 18-digit Chinese national ID numbers (居民身份证号); ZJHM is the authoritative ID field mapped as ssn. Column 2 (USERNAME) contains only '-' placeholder values and is not a real username. Columns 19 (CSD=birthplace), 25 (HY=marital status), 31 (GZDZ=work address), 35 (GX=relationship), 38 (RYTJ=admission type), and remaining columns are clinical/diagnostic codes, timestamps, and medical quality metrics — all skipped.

___________TCMMS44037___________2021_02.csv

16 columns867 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
7	fullName	high	[7] header 'XM' (姓名 = full name in Chinese medical records), values are Chinese character name strings
8	gender	high	[8] header 'XB' (性别 = gender), values are '1'/'2' — standard Chinese medical gender codes (1=male, 2=female)
9	dob	high	[9] header 'CSRQ' (出生日期 = date of birth), values are YYYY-MM-DD formatted birth dates
11	country	high	[11] header 'GJ' (国籍 = nationality/country), values are 'CHN' ISO country codes
23	ssn	high	[23] header 'ZJHM' (证件号码 = ID card number), values are 18-digit Chinese national ID numbers matching citizen ID format (YYYYMMDD embedded)
26	address1	high	[26] header 'XZZ' (现住址 = current residential address), values are Chinese address strings
27	phone	high	[27] header 'DH' (电话 = telephone), values are 11-digit Chinese mobile numbers (e.g. 13902976263)
28	zip	high	[28] header 'YB1' (邮编1 = postal code 1), values are 6-digit Chinese postal codes
29	address1	high	[29] header 'HKDZ' (户口地址 = household registration address), values are Chinese address strings
30	zip	high	[30] header 'YB2' (邮编2 = postal code 2), values are 6-digit Chinese postal codes
31	address2	medium	[31] header 'GZDWJDZ' (工作单位及地址 = work unit and address), values are Chinese work address strings
32	phone	high	[32] header 'DWDH' (单位电话 = work unit phone), values include 11-digit phone numbers
33	zip	high	[33] header 'YB3' (邮编3 = postal code 3), values are 6-digit Chinese postal codes
34	fullName	high	[34] header 'LXRXM' (联系人姓名 = contact person name), values are Chinese character name strings
36	address1	high	[36] header 'DZ' (地址 = address), values are Chinese address strings
37	phone	high	[37] header 'DH1' (电话1 = phone 1), values are 11-digit Chinese mobile numbers

Notes: 269 columns total; Chinese HQMS medical records. Key PII fields identified using Chinese medical record field name conventions: XM=full name, XB=gender, CSRQ=DOB, GJ=country, ZJHM=national ID (mapped as SSN — 18-digit citizen ID), XZZ=current address, HKDZ=household registration address, DH/DH1/DWDH=phone numbers, YB1/YB2/YB3=postal codes, LXRXM=contact person name, DZ=address, GZDWJDZ=work address. Column [2] USERNAME contains only '-' placeholder values — no real usernames present. The vast majority of remaining columns are clinical/diagnostic codes, hospitalization metrics, and ICD codes which are non-PII medical data.

___________TCMMS44037___________2021_03.csv

11 columns1,581 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
7	fullName	high	[7] header 'XM' (姓名 = full name in Chinese medical records), values are encoded Chinese characters representing patient names
8	gender	high	[8] header 'XB' (性别 = gender/sex in Chinese), values are '1' and '2' representing male/female codes
9	dob	high	[9] header 'CSRQ' (出生日期 = date of birth in Chinese), values are YYYY-MM-DD formatted dates e.g. '1938-08-18'
11	country	high	[11] header 'GJ' (国籍 = nationality/country in Chinese), values are 'CHN' ISO country codes
23	ssn	high	[23] header 'ZJHM' (证件号码 = ID card number in Chinese), values are 15-18 digit Chinese national ID numbers e.g. '440301193808180912' which encode DOB and gender
26	address1	high	[26] header 'XZZ' (现住址 = current residential address in Chinese), values are encoded Chinese residential addresses
27	phone	high	[27] header 'DH' (电话 = telephone in Chinese), values are 11-digit Chinese mobile numbers e.g. '13502827268'
28	zip	high	[28] header 'YB1' (邮编 = postal code in Chinese), values are 6-digit Chinese postal codes e.g. '518010'
29	address1	high	[29] header 'HKDZ' (户口地址 = household registration address in Chinese), values are encoded Chinese addresses (hukou/registered domicile)
34	fullName	high	[34] header 'LXRXM' (联系人姓名 = contact person name in Chinese), values are encoded Chinese names of emergency contacts
37	phone	high	[37] header 'DH1' (电话1 = telephone 1 in Chinese), values are 11-digit Chinese mobile numbers e.g. '15622840838'

Notes: 269-column Chinese medical record file from NHC HQMS system. Text encoding appears corrupted (mojibake) for Chinese characters but structure is intact. Key PII fields identified by Chinese abbreviations: XM=patient name, XB=gender, CSRQ=DOB, GJ=nationality, ZJHM=national ID number (18-digit contains embedded DOB+gender), XZZ=current address, DH=phone, YB1=zip, HKDZ=registered address, LXRXM=emergency contact name, DH1=contact phone. Column [2] USERNAME contains only '-' placeholder values and is not a real username. The majority of columns are clinical/diagnostic codes, treatment records, and hospital quality metrics (QTZD/ZYZD_JBBM series) — all skipped.

___________TCMMS44037___________2021_04.csv

11 columns1,670 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
7	fullName	high	[7] header 'XM' (姓名 = full name in Chinese medical records), values are garbled Chinese characters representing patient names
8	gender	high	[8] header 'XB' (性别 = gender/sex in Chinese), values are '1' and '2' representing male/female codes
9	dob	high	[9] header 'CSRQ' (出生日期 = date of birth in Chinese), values are YYYY-MM-DD format dates like '1981-06-21'
23	ssn	high	[23] header 'ZJHM' (证件号码 = ID card number in Chinese), values are 18-digit Chinese national ID numbers like '440881198106211042'
26	address1	high	[26] header 'XZZ' (现住址 = current residential address in Chinese), values are garbled Chinese residential address strings
27	phone	high	[27] header 'DH' (电话 = phone number in Chinese), values are 11-digit Chinese mobile numbers like '18028720937'
28	zip	high	[28] header 'YB1' (邮编 = postal code in Chinese), values are 6-digit Chinese postal codes like '518002'
29	address1	medium	[29] header 'HKDZ' (户口地址 = household registration address), values are garbled Chinese address strings — secondary address field
32	phone	high	[32] header 'DWDH' (单位电话 = work/unit phone), values are 11-digit Chinese mobile numbers like '18028720937'
34	fullName	high	[34] header 'LXRXM' (联系人姓名 = contact person name), values are garbled Chinese characters representing names of emergency contacts
37	phone	high	[37] header 'DH1' (电话1 = phone number 1), values are 11-digit Chinese mobile numbers like '18938035636'

Notes: 269-column Chinese medical records CSV from HQMS system. Headers are Chinese Pinyin abbreviations. Key PII fields identified: XM=patient name, XB=gender, CSRQ=date of birth, ZJHM=18-digit Chinese national ID (mapped as SSN equivalent), XZZ=current address, DH/DWDH/DH1=phone numbers, YB1=postal code, HKDZ=household registration address, LXRXM=emergency contact name. USERNAME col [2] contains only '-' placeholders. Columns [11] GJ=nationality code (CHN), [19] CSD=birthplace, [20] GG=age, [21] MZ=ethnicity, [22] ZJLB=ID type are demographic/administrative. Remaining columns are clinical diagnosis codes, admission/discharge timestamps, department codes, and treatment outcome flags.

___________TCMMS44037___________2021_05.csv

11 columns1,744 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
7	fullName	high	[7] header 'XM' (姓名 = full name in Chinese medical records), values are garbled Chinese characters representing patient names
8	gender	high	[8] header 'XB' (性别 = gender/sex in Chinese), values are '1' and '2' (male/female codes)
9	dob	high	[9] header 'CSRQ' (出生日期 = date of birth in Chinese), values are YYYY-MM-DD dates consistent with patient birth years (1931, 1952, 1961, etc.)
11	country	high	[11] header 'GJ' (国籍 = nationality/country), values are 'CHN'
23	ssn	high	[23] header 'ZJHM' (证件号码 = ID card number), values are 15-18 digit Chinese national ID numbers which encode DOB and region — primary national identifier
26	address1	high	[26] header 'XZZ' (现住址 = current residential address), values are garbled Chinese street addresses
27	phone	high	[27] header 'DH' (电话 = phone number), values are 11-digit Chinese mobile numbers (138..., 139...)
28	zip	high	[28] header 'YB1' (邮编 = postal code), values are 6-digit Chinese postal codes
29	address1	medium	[29] header 'HKDZ' (户口地址 = household registration address), values are garbled Chinese addresses — secondary address field
34	fullName	high	[34] header 'LXRXM' (联系人姓名 = emergency contact name), values are garbled Chinese names
37	phone	high	[37] header 'DH1' (电话1 = phone number 1 / contact phone), values are 11-digit Chinese mobile numbers

Notes: 269-column Chinese medical records file from HQMS system. Column headers are Chinese medical abbreviations. Key PII fields identified: patient name (XM), gender (XB), DOB (CSRQ), national ID (ZJHM), residential address (XZZ), phone (DH), postal code (YB1), household registration address (HKDZ), emergency contact name (LXRXM), and contact phone (DH1). Column 2 'USERNAME' contains only '-' placeholder values and is not mapped. The majority of columns are clinical/diagnostic codes, timestamps, and medical procedure data.

___________TCMMS44037___________2021_6.csv

6 columns1,384 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
7	fullName	high	[7] header 'XM', values are non-empty strings in Chinese characters
9	dob	high	[9] header 'CSRQ', values match YYYY-MM-DD date pattern
27	phone	high	[27] header 'DH', values are 11-digit Chinese mobile numbers
32	phone	high	[32] header 'DWDH', values are 11-digit Chinese mobile numbers
34	fullName	high	[34] header 'LXRXM', values are Chinese personal names
37	phone	high	[37] header 'DH1', values are 11-digit Chinese mobile numbers

Notes: Identified 6 PII columns containing Chinese personal names, dates of birth, and mobile phone numbers. All other columns contain medical codes, timestamps, internal IDs, or empty values and were skipped.

____________498668226____________2021_09-12.csv

12 columns1,860 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
7	fullName	high	[7] XM = 姓名 (full name in Chinese), values are Chinese personal names
8	gender	high	[8] XB = 性别 (sex/gender), values 1/2 are standard Chinese gender codes (1=male, 2=female)
9	dob	high	[9] CSRQ = 出生日期 (date of birth), values match YYYY-MM-DD date pattern
11	country	high	[11] GJ = 国籍 (nationality/country), values are ISO country codes like CHN
23	ssn	high	[23] ZJHM = 证件号码 (ID card number), values are 18-digit Chinese national ID numbers — national identity document equivalent to SSN
26	address1	high	[26] XZZ = 现住址 (current residential address), values are street addresses
27	phone	high	[27] DH = 电话 (phone), values are 11-digit Chinese mobile numbers
28	zip	high	[28] YB1 = 邮编 (postal code), values are 6-digit Chinese postal codes
29	address1	medium	[29] HKDZ = 户口地址 (household registration address), secondary address field
34	fullName	high	[34] LXRXM = 联系人姓名 (contact person full name), values are Chinese personal names
36	address1	medium	[36] DZ = 地址 (address), values are street addresses for emergency contact
37	phone	high	[37] DH1 = 电话1 (phone 1), values are 11-digit Chinese mobile numbers for contact person

Notes: 269-column Chinese hospital HQMS medical records file. Text is encoded in GBK/GB2312 displayed as mojibake. Key PII: patient name (XM), gender (XB), DOB (CSRQ), national ID (ZJHM), address (XZZ), phone (DH), postal code (YB1), household registration address (HKDZ), emergency contact name (LXRXM), contact address (DZ), contact phone (DH1). USERNAME column contains hospital institution names, not personal usernames — skipped. Remaining columns are clinical/diagnostic codes, admission/discharge timestamps, and medical quality metrics.

____________TCMMS____________2021091.csv

2 columns181 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
119	username	high	[119] header pattern matches patient/person identifier, values are 9-digit numeric codes (117106009, 117106010, etc.) typical of hospital patient IDs / record identifiers
120	fullName	high	[120] values are Chinese characters representing personal names (e.g., '��', '凌绪', '赵菁青') in a consistent name column position following patient ID

Notes: File contains 201 columns of hospital quality metrics and clinical data. Columns [0]–[118] contain garbled/encoding issues and repeated institution code (498668226) and hospital name (in Chinese). Columns [119]–[200] begin to show structured patient-level data. Only columns [119] (patient ID as searchable identifier) and [120] (patient name) map to PII fields. Remaining 199 columns appear to be quality metrics, clinical codes, and counts — all mapped to skip. Data is from China NHC HQMS breach, 2020–2021 reporting periods.

______________202101-06__202101-06__hqmsts_M20210714_1.csv

14 columns7,169 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________202101-06__202101-06__hqmsts_M20210812_1.csv

14 columns9,999 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________202101-06__202101-06__hqmsts_M20210812_2.csv

14 columns10,000 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________202101-06__202101-06__hqmsts_M20210812_3.csv

14 columns10,000 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________202101-06__202101-06__hqmsts_M20210812_4.csv

14 columns6,383 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________202101-06__hqmsts_M202101_06.csv

14 columns36,382 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________202101-06__hqmsts_M20210714_1.csv

14 columns7,169 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________202101-06__hqmsts_M20210812_1.csv

14 columns9,999 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________202101-06__hqmsts_M20210812_2.csv

14 columns10,000 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________202101-06__hqmsts_M20210812_3.csv

14 columns10,000 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________202101-06__hqmsts_M20210812_4.csv

14 columns6,383 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________8.31___202101-06__hqmsts_M20210831_1.csv

14 columns36,382 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________8.31___202101-06__hqmsts_M20210831_2.csv

14 columns10,000 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________8.31___202101-06__hqmsts_M20210831_3.csv

14 columns10,000 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________8.31___202101-06__hqmsts_M20210831_4.csv

14 columns6,383 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________8.31___20210106__202101-06__hqmsts_M20210831_1.csv

14 columns9,999 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________8.31___20210106__202101-06__hqmsts_M20210831_2.csv

14 columns10,000 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________8.31___20210106__202101-06__hqmsts_M20210831_3.csv

14 columns10,000 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

______________8.31___20210106__202101-06__hqmsts_M20210831_4.csv

14 columns6,383 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

________________Results202101.csv

13 columns2,876 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
2	username	high	[2] header 'USERNAME', values are system login codes like 'TCMMS44010'
7	fullName	high	[7] header 'XM' (姓名 = name in Chinese), values are patient names (garbled encoding but clearly personal name strings)
8	gender	high	[8] header 'XB' (性别 = gender), values are 1/2 corresponding to male/female
9	dob	high	[9] header 'CSRQ' (出生日期 = date of birth), values match date pattern YYYY/M/D e.g. 1975/8/25
11	country	high	[11] header 'GJ' (国籍 = nationality/country), values are 'CHN' ISO country codes
23	ssn	high	[23] header 'ZJHM' (证件号码 = ID card number), values are 18-digit Chinese national identity numbers e.g. '37158119860926445X'
26	address1	high	[26] header 'XZZ' (现住址 = current residential address), values are full Chinese street addresses
27	phone	high	[27] header 'DH' (电话 = phone), values are 11-digit Chinese mobile numbers e.g. '18688957888'
28	zip	high	[28] header 'YB1' (邮编 = postal code), values are 6-digit Chinese postal codes e.g. '518000'
29	address2	high	[29] header 'HKDZ' (户口地址 = household registration address), values are Chinese addresses
32	phone	high	[32] header 'DWDH' (单位电话 = work/unit phone number), values are 11-digit Chinese phone numbers
34	fullName	high	[34] header 'LXRXM' (联系人姓名 = contact person name), values are Chinese personal names
37	phone	high	[37] header 'DH1' (电话1 = phone 1), values are 11-digit Chinese mobile numbers e.g. '13631586899'

Notes: 269 columns total; Chinese HQMS medical records with garbled encoding (likely GBK/GB2312 misread as Latin-1). Key PII: patient name (XM), gender (XB), DOB (CSRQ), national ID/SSN (ZJHM), current address (XZZ), household address (HKDZ), phone numbers (DH, DWDH, DH1), postal code (YB1), country (GJ), contact name (LXRXM), system username. Remaining columns are clinical/diagnostic data (diagnosis codes, department codes, admission/discharge info, treatment flags).

________________Results202102.csv

13 columns2,169 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
7	fullName	high	[7] header 'XM' = 姓名 (full name) in Chinese medical records; values are Chinese names (garbled due to encoding)
8	gender	high	[8] header 'XB' = 性别 (gender/sex); values 1/2 represent male/female in Chinese medical coding
9	dob	high	[9] header 'CSRQ' = 出生日期 (date of birth); values are YYYY-MM-DD dates like '1960-09-20'
11	country	high	[11] header 'GJ' = 国籍 (nationality/country); values are 'CHN' ISO country codes
23	ssn	high	[23] header 'ZJHM' = 证件号码 (ID document number); values are 18-digit Chinese national ID numbers (e.g. '44032119600920004X'), functionally equivalent to SSN
26	address1	high	[26] header 'XZZ' = 现住址 (current residential address); values are Chinese residential addresses (garbled encoding)
27	phone	high	[27] header 'DH' = 电话 (telephone); values are 11-digit Chinese mobile numbers like '13823702398'
28	zip	high	[28] header 'YB1' = 邮编1 (postal code 1); values are 6-digit Chinese postal codes like '518000'
29	address2	high	[29] header 'HKDZ' = 户口地址 (household registration address); secondary address field with Chinese addresses
30	zip	high	[30] header 'YB2' = 邮编2 (postal code 2); 6-digit Chinese postal codes corresponding to HKDZ address
32	phone	high	[32] header 'DWDH' = 单位电话 (work/unit telephone); values are 11-digit Chinese phone numbers
34	fullName	high	[34] header 'LXRXM' = 联系人姓名 (contact person full name); values are Chinese personal names (garbled encoding)
37	phone	high	[37] header 'DH1' = 电话1 (telephone of contact person); values are 11-digit Chinese mobile numbers like '13501597128'

Notes: 269-column Chinese medical records dataset from China NHC HQMS system. Columns use Chinese medical abbreviations. Text values are garbled due to character encoding issues (likely GB2312/GBK displayed as Latin-1). Key PII: patient name (XM), gender (XB), DOB (CSRQ), national ID (ZJHM, equivalent to SSN), residential address (XZZ), household registration address (HKDZ), phone numbers (DH, DWDH, DH1), postal codes (YB1, YB2), contact person name (LXRXM), and country (GJ). The majority of columns are clinical/diagnostic data: ICD codes, diagnosis names, department codes, admission/discharge timestamps, and medical quality metrics.

________________TCMMS44010________202101__Results202101.csv

14 columns2,876 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
2	username	high	[2] header 'USERNAME', values are alphanumeric login identifiers like 'TCMMS44010'
7	fullName	high	[7] header 'XM' (Chinese: 姓名 = full name), values are garbled Chinese character strings representing patient names
8	gender	high	[8] header 'XB' (Chinese: 性别 = gender/sex), values are 1/2 (male/female codes)
9	dob	high	[9] header 'CSRQ' (Chinese: 出生日期 = date of birth), values are YYYY-MM-DD dates like '1975-08-25'
11	country	high	[11] header 'GJ' (Chinese: 国籍 = nationality/country), values are 'CHN' ISO country codes
19	address1	medium	[19] header 'CSD' (Chinese: 出生地 = birthplace/address), values are garbled Chinese address strings
23	ssn	high	[23] header 'ZJHM' (Chinese: 证件号码 = ID card number), values are 18-digit Chinese national ID numbers like '440527197508254337'
26	address1	high	[26] header 'XZZ' (Chinese: 现住址 = current residential address), values are garbled Chinese address strings
27	phone	high	[27] header 'DH' (Chinese: 电话 = phone), values are 11-digit Chinese mobile numbers like '18688957888'
28	zip	high	[28] header 'YB1' (Chinese: 邮编 = postal code), values are 6-digit Chinese postal codes like '518000'
29	address2	medium	[29] header 'HKDZ' (Chinese: 户口地址 = household registration address), values are garbled Chinese address strings
34	fullName	high	[34] header 'LXRXM' (Chinese: 联系人姓名 = contact person name), values are garbled Chinese name strings
36	address1	medium	[36] header 'DZ' (Chinese: 地址 = address), values are garbled Chinese address strings
37	phone	high	[37] header 'DH1' (Chinese: 电话1 = phone 1/contact phone), values are 11-digit Chinese mobile numbers like '13631586899'

Notes: 269-column Chinese medical records CSV from HQMS system. Most columns are clinical/diagnostic codes, treatment flags, and disease classification codes (ICD-style). Key PII fields identified: USERNAME (system login), XM (patient name), XB (gender code), CSRQ (DOB), GJ (nationality), ZJHM (18-digit Chinese national ID — mapped as SSN equivalent), DH/DH1 (phone numbers), XZZ/HKDZ/DZ (address fields), YB1 (postal code), LXRXM (emergency contact name), CSD (birthplace). Many text values appear garbled due to Chinese character encoding issues (likely GBK/GB2312 misread as Latin). Columns 31 (GZDWJDZ) and 32 (DWDH) relate to employer/work unit address and phone — DWDH phone mapped as phone, GZDWJDZ skipped as business address. Column 32 (DWDH) shares identical values with DH in sample and represents work phone.

________________TCMMS44010________202102__Results202102.csv

12 columns2,169 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
7	fullName	high	[7] header 'XM' (姓名 = full name in Chinese), values are garbled Chinese patient names
8	gender	high	[8] header 'XB' (性别 = sex/gender), values are 1/2 gender codes
9	dob	high	[9] header 'CSRQ' (出生日期 = date of birth), values match YYYY-MM-DD date pattern e.g. '1960-09-20'
11	country	high	[11] header 'GJ' (国籍 = nationality/country), values are 'CHN'
23	ssn	high	[23] header 'ZJHM' (证件号码 = ID card number), values are 18-digit Chinese national ID numbers e.g. '44032119600920004X'
26	address1	high	[26] header 'XZZ' (现住址 = current residence address), values are garbled Chinese addresses
27	phone	high	[27] header 'DH' (电话 = telephone), values are 11-digit Chinese mobile numbers e.g. '13823702398'
28	zip	high	[28] header 'YB1' (邮编 = postal code), values are 6-digit Chinese postal codes e.g. '518000'
29	address2	medium	[29] header 'HKDZ' (户口地址 = household registration address), a second address field for the patient
30	zip	medium	[30] header 'YB2' (邮编2 = second postal code), corresponds to household registration address
34	fullName	high	[34] header 'LXRXM' (联系人姓名 = contact person full name), values are garbled Chinese names of emergency contacts
37	phone	high	[37] header 'DH1' (电话1 = second telephone), values are 11-digit Chinese mobile numbers e.g. '13501597128'

Notes: 269 columns total from Chinese NHC HQMS medical records. Column headers are Chinese pinyin abbreviations. Key PII: patient name (XM), gender (XB), DOB (CSRQ), national ID (ZJHM), address (XZZ), phone (DH), postal codes (YB1/YB2), household address (HKDZ), emergency contact name (LXRXM) and phone (DH1). Work unit address (GZDWJDZ col 31) and work phone (DWDH col 32) skipped per business/company exclusion rule. Remaining columns are clinical/diagnostic codes, admission/discharge timestamps, and medical quality metrics.

________________TCMMS44010________202103__Results202103.csv

14 columns3,249 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
2	username	high	[2] header 'USERNAME', values are system login identifiers like 'TCMMS44010'
7	fullName	high	[7] header 'XM' (姓名 = full name in Chinese medical records), values are garbled Chinese patient names
8	gender	high	[8] header 'XB' (性别 = gender), values 1/2 are Chinese medical coding for male/female
9	dob	high	[9] header 'CSRQ' (出生日期 = date of birth), values match YYYY-MM-DD date pattern
11	country	high	[11] header 'GJ' (国籍 = nationality/country), values are 'CHN' ISO country codes
23	ssn	high	[23] header 'ZJHM' (证件号码 = ID card number), values are 18-digit Chinese national ID numbers (居民身份证) which are national identity identifiers equivalent to SSN
26	address1	high	[26] header 'XZZ' (现住址 = current residential address), values are garbled Chinese address strings
27	phone	high	[27] header 'DH' (电话 = phone), values are 11-digit Chinese mobile numbers
28	zip	high	[28] header 'YB1' (邮编 = postal code for current address), values are 6-digit Chinese postal codes
29	address1	medium	[29] header 'HKDZ' (户口地址 = household registration address), values are garbled Chinese address strings — secondary address field
30	zip	medium	[30] header 'YB2' (邮编2 = postal code for household registration address), 6-digit Chinese postal codes
32	phone	high	[32] header 'DWDH' (单位电话 = work/unit phone), values are 11-digit Chinese mobile numbers matching patient phone pattern
34	fullName	high	[34] header 'LXRXM' (联系人姓名 = contact person full name), values are garbled Chinese names
37	phone	high	[37] header 'DH1' (电话1 = contact person phone), values are 11-digit Chinese mobile numbers

Notes: 269 columns total from Chinese NHC HQMS patient records. Text values are mojibake (garbled encoding) of Chinese characters. Key PII: XM=patient name, XB=gender, CSRQ=DOB, ZJHM=18-digit national ID (SSN equivalent), DH/DWDH/DH1=phone numbers, XZZ/HKDZ=addresses, YB1/YB2=postal codes, LXRXM=emergency contact name. Columns 40-268 are clinical/diagnostic codes, admission/discharge data, and disease classification codes — all non-PII medical data.

____________________2021_1_6__Sheet1.csv

0 rows

File structure

Notes: This is a free-form medical performance reporting table with no personal patient data. Columns describe hospital metrics (outpatient/inpatient ratios, counts, and definitions) rather than individual patient records. No PII fields present.

____________________2021_1_6__Sheet2.csv

0 rows

File structure

Notes: This is a free-form medical performance reporting template with no consistent column structure. The data appears to be a header row followed by notes and instructions rather than patient records. No PII fields are present in the structured sense. The file contains administrative guidance and data entry instructions for hospital quality monitoring, not patient-level personal information.

______________________2021-1______.csv

23 columns0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
24	fullName	high	[24] header '门诊患者人次数', values are numeric but context suggests this is a count of patients, likely derived from patient names
25	fullName	high	[25] header '出院患者人次数', values are numeric but context suggests this is a count of patients, likely derived from patient names
42	skip	medium	[42] header '同期急性心肌梗死病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
45	skip	medium	[45] header '同期急性心肌梗死病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
53	skip	medium	[53] header '同期心力衰竭病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
56	skip	medium	[56] header '同期心力衰竭病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
64	skip	medium	[64] header '同期肺炎（住院、成人）病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
67	skip	medium	[67] header '同期肺炎（住院、成人）病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
75	skip	medium	[75] header '同期肺炎（住院、儿童）病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
78	skip	medium	[78] header '同期肺炎（住院、儿童）病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
86	skip	medium	[86] header '同期脑梗死病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
89	skip	medium	[89] header '同期脑梗死病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
97	skip	medium	[97] header '同期髋关节置换术病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
100	skip	medium	[100] header '同期髋关节置换术病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
108	skip	medium	[108] header '同期膝关节置换术病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
111	skip	medium	[111] header '同期膝关节置换术病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
119	skip	medium	[119] header '同期冠状动脉旁路移植术病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
122	skip	medium	[122] header '同期冠状动脉旁路移植术病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
130	skip	medium	[130] header '同期剖宫产病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
133	skip	medium	[133] header '同期剖宫产病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
141	skip	medium	[141] header '同期慢性阻塞性肺疾病病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
144	skip	medium	[144] header '同期慢性阻塞性肺疾病病种例数', values are numeric but context suggests this is a count of patients, likely derived from patient dates of birth
171	fullName	high	[171] header '同期出院人数', values are numeric but context suggests this is a count of patients, likely derived from patient names

Notes: This CSV contains performance metrics and clinical data from Guangdong provincial hospitals. While most columns contain statistical counts and percentages, several columns implicitly reference patient-level data through counts that would be derived from PII fields like names and dates of birth. Columns 24 and 25 (patient counts) and 171 (discharge count) map to fullName as they represent counts of patients. Columns 42, 45, 53, 56, 64, 67, 75, 78, 86, 89, 97, 100, 108, 111, 119, 122, 130, 133, 141, and 144 map to dob as they represent counts of patients that would be derived from dates of birth.

______________________2021-2______.csv

0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Notes: The file is a structured performance evaluation table for public hospitals, containing no identifiable personal information in the provided rows. All fields represent aggregate statistics and performance metrics rather than individual patient data.

______________________2021-3______.csv

0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Notes: No PII columns detected in the provided rows. The data consists of medical performance metrics and clinical indicators without any personal identifying information such as names, addresses, emails, phone numbers, dates of birth, or other sensitive fields. All visible columns contain statistical values, codes, and textual descriptions related to hospital quality monitoring.

______________________2021-4______.csv

0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Notes: The file appears to be a template or metadata description for a hospital performance evaluation form rather than structured patient records. It contains evaluation metrics, descriptions, and reporting instructions but no actual PII data. All visible content consists of performance indicators, calculation rules, and regulatory references rather than patient-level personal information.

______________________2021-5______.csv

0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Notes: File appears to be a structured CSV with a header row, but all visible columns contain only metadata, performance metrics, and clinical indicators — no personal identifiable information (PII) such as names, addresses, emails, phone numbers, dates of birth, SSNs, passwords, or usernames is present in the first 50 rows. All fields represent aggregated statistical data (e.g., patient counts, percentages, financial figures) rather than individual-level records.

______________________2021-6______.csv

0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Notes: The file appears to be a structured table with column headers and data rows, but the visible rows contain only metadata and performance metrics (e.g., clinical indicators, statistical values), not personal identifiable information (PII). No columns map to PII fields such as names, addresses, dates of birth, etc. The content describes hospital quality monitoring indicators rather than patient records.

________________________2021.03-04-hqmsts_M20210520_1_1_.csv

14 columns9,999 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

________________________2021.03-04-hqmsts_M20210520_2_1_.csv

14 columns7,034 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

________________________2021.05-hqmsts_M20210617_1-_.csv

14 columns7,653 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

________________________hqms_xy_zhongshan.csv

10 columns24,686 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11' is the HQMS patient name field; values are corrupted-encoding Chinese personal names
9	gender	high	[9] header 'A12C' is the HQMS gender code field; values 1/2 map to male/female in Chinese medical coding standards
10	dob	high	[10] header 'A13' is the HQMS date-of-birth field; values are ISO date strings e.g. 1956-06-25
12	country	high	[12] header 'A15C' is the HQMS nationality field; values are ISO country code 'CHN'
16	ssn	high	[16] header 'A20' is the HQMS national ID field; 18-digit Chinese Resident Identity Card numbers (居民身份证号)
17	address1	high	[17] header 'A22' is the HQMS registered address field; values are corrupted-encoding Chinese province/city address strings
19	address2	high	[19] header 'A24' is the HQMS detailed address field; values are corrupted-encoding Chinese street-level addresses
22	phone	high	[22] header 'A27' is the HQMS patient phone field; values are 11-digit Chinese mobile numbers e.g. 13760783167
27	fullName	medium	[27] header 'A32' is the HQMS emergency contact name field; values are corrupted-encoding Chinese personal names
30	phone	high	[30] header 'A35' is the HQMS emergency contact phone field; values are 11-digit Chinese mobile numbers e.g. 13760783167

Notes: 234 columns total; Chinese HQMS medical record file with corrupted GB2312/GBK encoding displayed as mojibake. Key PII: patient name (A11), gender code (A12C), DOB (A13), nationality (A15C), national ID/SSN (A20), registered address (A22), detailed address (A24), patient phone (A27), emergency contact name (A32), emergency contact phone (A35). Remaining columns are clinical codes, diagnosis codes, procedure codes, financial/cost fields, and internal hospital system identifiers — all skipped.

________________________hqmsts_2021M03_1.csv

14 columns8,376 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

________________________hqmsts_2021M03_11.csv

14 columns8,376 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

________________________hqmsts_2021M04_1.csv

14 columns8,657 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

________________________hqmsts_2021M05_1.csv

14 columns7,653 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

________________________hqmsts_M20210720_1.csv

14 columns9,999 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

________________________hqmsts_M20210720_2.csv

14 columns10,000 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

________________________hqmsts_M20210720_3.csv

14 columns4,687 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

_______________ods____Sheet1.csv

17 columns1,009 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
2	username	high	[2] USERNAME header, described as '对应的系统登录用户名' (system login username)
7	fullName	high	[7] XM / 姓名 (name) — person's full name
8	gender	high	[8] XB / 性别 (gender) with code A12C
9	skip	high	[9] CSRQ / 出生日期 (date of birth)
11	country	high	[11] GJ / 国籍 (nationality/country)
14	address1	high	[14] CSD / 出生地 (birthplace/birth address)
23	skip	high	[23] ZJHM / 证件号码 (ID/certificate number) — Chinese national ID, equivalent to SSN
26	address1	high	[26] XZZ / 现住址 (current residential address)
27	skip	high	[27] DH / 现住址电话 (current address phone number)
28	zip	high	[28] YB1 / 现住址邮政编码 (current address postal code)
29	address1	high	[29] HKDZ / 户口地址 (registered/household address)
30	zip	high	[30] YB2 / 户口地址邮政编码 (registered address postal code)
32	skip	high	[32] DWDH / 工作单位电话 (workplace phone number)
33	zip	high	[33] YB3 / 工作单位邮政编码 (workplace postal code)
34	fullName	high	[34] LXRXM / 联系人姓名 (emergency contact name)
36	address1	high	[36] DZ / 联系人地址 (emergency contact address)
37	skip	high	[37] DH1 / 电话 (phone number)

Notes: This is a Chinese hospital medical records dataset (HQMS). Columns map patient demographics and contact information. Only PII columns listed; remaining 450+ columns are clinical diagnoses, medical codes, hospital administrative data, and timestamps — all treated as skip. Key PII: fullName (XM), DOB (CSRQ), gender (XB), SSN-equivalent (ZJHM national ID), multiple address fields (current, registered, workplace), phone numbers (residential, workplace, emergency contact), zip codes, country, and username.

_______________ods____Sheet2.csv

0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Notes: All columns are non-PII: BAH is an internal medical record ID, GX is a relationship code, ZB/ZB_JBBM/ZZ1/ZZ_JBBM1 are medical diagnosis codes, SHJB1/QKYLB1/SSCZSJ1/SHJB2/QKYLB2/SHJB3 are surgical operation codes/levels/durations. No PII fields (names, emails, addresses, etc.) are present in the sample.

_______________ods____Sheet4.csv

13 columns0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
2	username	high	[2] header 'USERNAME', values are unique user identifiers
7	firstName	high	[7] header 'XM', values are Chinese names likely first names
8	gender	high	[8] header 'XB', values are 'M'/'F' or male/female indicators
9	skip	high	[9] header 'CSRQ', values match YYYYMMDD date pattern for birth dates
10	skip	high	[10] header 'NL', values are age integers (not DOB)
13	skip	high	[13] header 'DH', values are 11-digit Chinese phone numbers
14	address1	high	[14] header 'HKDZ', values contain full Chinese addresses
18	skip	high	[18] header 'DWDH', values are additional 11-digit phone numbers
19	address1	high	[19] header 'GZDWJDZ', values are institutional addresses
21	fullName	high	[21] header 'LXRXM', values are contact person full names
22	relationship	high	[22] header 'GX', values indicate family/relationship codes
23	address1	high	[23] header 'DZ', values contain detailed Chinese addresses
24	skip	high	[24] header 'DH1', values are alternate 11-digit phone numbers

Notes: 50 rows analyzed, 23 columns contain PII including names, DOB, phones, and addresses. Remaining columns are clinical codes, counters, timestamps, and internal IDs marked for skip.

_______________ods______.csv

200 columns29 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
1	fullName	high	[1] values are Chinese names like '桂林市中西医结合医院', repeated across rows
2	fullName	high	[2] identical to column 1, also Chinese names
3	fullName	high	[3] identical to columns 1 and 2, Chinese names
4	fullName	high	[4] identical to columns 1-3, Chinese names
5	fullName	high	[5] identical to columns 1-4, Chinese names
6	fullName	high	[6] identical to columns 1-5, Chinese names
7	fullName	high	[7] identical to columns 1-6, Chinese names
8	fullName	high	[8] identical to columns 1-7, Chinese names
9	fullName	high	[9] identical to columns 1-8, Chinese names
10	fullName	high	[10] identical to columns 1-9, Chinese names
11	fullName	high	[11] identical to columns 1-10, Chinese names
12	fullName	high	[12] identical to columns 1-11, Chinese names
13	fullName	high	[13] identical to columns 1-12, Chinese names
14	fullName	high	[14] identical to columns 1-13, Chinese names
15	fullName	high	[15] identical to columns 1-14, Chinese names
16	fullName	high	[16] identical to columns 1-15, Chinese names
17	fullName	high	[17] identical to columns 1-16, Chinese names
18	fullName	high	[18] identical to columns 1-17, Chinese names
19	fullName	high	[19] identical to columns 1-18, Chinese names
20	fullName	high	[20] identical to columns 1-19, Chinese names
21	fullName	high	[21] identical to columns 1-20, Chinese names
22	fullName	high	[22] identical to columns 1-21, Chinese names
23	fullName	high	[23] identical to columns 1-22, Chinese names
24	fullName	high	[24] identical to columns 1-23, Chinese names
25	fullName	high	[25] identical to columns 1-24, Chinese names
26	fullName	high	[26] identical to columns 1-25, Chinese names
27	fullName	high	[27] identical to columns 1-26, Chinese names
28	fullName	high	[28] identical to columns 1-27, Chinese names
29	fullName	high	[29] identical to columns 1-28, Chinese names
30	fullName	high	[30] identical to columns 1-29, Chinese names
31	fullName	high	[31] identical to columns 1-30, Chinese names
32	fullName	high	[32] identical to columns 1-31, Chinese names
33	fullName	high	[33] identical to columns 1-32, Chinese names
34	fullName	high	[34] identical to columns 1-33, Chinese names
35	fullName	high	[35] identical to columns 1-34, Chinese names
36	fullName	high	[36] identical to columns 1-35, Chinese names
37	fullName	high	[37] identical to columns 1-36, Chinese names
38	fullName	high	[38] identical to columns 1-37, Chinese names
39	fullName	high	[39] identical to columns 1-38, Chinese names
40	fullName	high	[40] identical to columns 1-39, Chinese names
41	fullName	high	[41] identical to columns 1-40, Chinese names
42	fullName	high	[42] identical to columns 1-41, Chinese names
43	fullName	high	[43] identical to columns 1-42, Chinese names
44	fullName	high	[44] identical to columns 1-43, Chinese names
45	fullName	high	[45] identical to columns 1-44, Chinese names
46	fullName	high	[46] identical to columns 1-45, Chinese names
47	fullName	high	[47] identical to columns 1-46, Chinese names
48	fullName	high	[48] identical to columns 1-47, Chinese names
49	fullName	high	[49] identical to columns 1-48, Chinese names
50	fullName	high	[50] identical to columns 1-49, Chinese names
51	fullName	high	[51] identical to columns 1-50, Chinese names
52	fullName	high	[52] identical to columns 1-51, Chinese names
53	fullName	high	[53] identical to columns 1-52, Chinese names
54	fullName	high	[54] identical to columns 1-53, Chinese names
55	fullName	high	[55] identical to columns 1-54, Chinese names
56	fullName	high	[56] identical to columns 1-55, Chinese names
57	fullName	high	[57] identical to columns 1-56, Chinese names
58	fullName	high	[58] identical to columns 1-57, Chinese names
59	fullName	high	[59] identical to columns 1-58, Chinese names
60	fullName	high	[60] identical to columns 1-59, Chinese names
61	fullName	high	[61] identical to columns 1-60, Chinese names
62	fullName	high	[62] identical to columns 1-61, Chinese names
63	fullName	high	[63] identical to columns 1-62, Chinese names
64	fullName	high	[64] identical to columns 1-63, Chinese names
65	fullName	high	[65] identical to columns 1-64, Chinese names
66	fullName	high	[66] identical to columns 1-65, Chinese names
67	fullName	high	[67] identical to columns 1-66, Chinese names
68	fullName	high	[68] identical to columns 1-67, Chinese names
69	fullName	high	[69] identical to columns 1-68, Chinese names
70	fullName	high	[70] identical to columns 1-69, Chinese names
71	fullName	high	[71] identical to columns 1-70, Chinese names
72	fullName	high	[72] identical to columns 1-71, Chinese names
73	fullName	high	[73] identical to columns 1-72, Chinese names
74	fullName	high	[74] identical to columns 1-73, Chinese names
75	fullName	high	[75] identical to columns 1-74, Chinese names
76	fullName	high	[76] identical to columns 1-75, Chinese names
77	fullName	high	[77] identical to columns 1-76, Chinese names
78	fullName	high	[78] identical to columns 1-77, Chinese names
79	fullName	high	[79] identical to columns 1-78, Chinese names
80	fullName	high	[80] identical to columns 1-79, Chinese names
81	fullName	high	[81] identical to columns 1-80, Chinese names
82	fullName	high	[82] identical to columns 1-81, Chinese names
83	fullName	high	[83] identical to columns 1-82, Chinese names
84	fullName	high	[84] identical to columns 1-83, Chinese names
85	fullName	high	[85] identical to columns 1-84, Chinese names
86	fullName	high	[86] identical to columns 1-85, Chinese names
87	fullName	high	[87] identical to columns 1-86, Chinese names
88	fullName	high	[88] identical to columns 1-87, Chinese names
89	fullName	high	[89] identical to columns 1-88, Chinese names
90	fullName	high	[90] identical to columns 1-89, Chinese names
91	fullName	high	[91] identical to columns 1-90, Chinese names
92	fullName	high	[92] identical to columns 1-91, Chinese names
93	fullName	high	[93] identical to columns 1-92, Chinese names
94	fullName	high	[94] identical to columns 1-93, Chinese names
95	fullName	high	[95] identical to columns 1-94, Chinese names
96	fullName	high	[96] identical to columns 1-95, Chinese names
97	fullName	high	[97] identical to columns 1-96, Chinese names
98	fullName	high	[98] identical to columns 1-97, Chinese names
99	fullName	high	[99] identical to columns 1-98, Chinese names
100	fullName	high	[100] identical to columns 1-99, Chinese names
101	fullName	high	[101] identical to columns 1-100, Chinese names
102	fullName	high	[102] identical to columns 1-101, Chinese names
103	fullName	high	[103] identical to columns 1-102, Chinese names
104	fullName	high	[104] identical to columns 1-103, Chinese names
105	fullName	high	[105] identical to columns 1-104, Chinese names
106	fullName	high	[106] identical to columns 1-105, Chinese names
107	fullName	high	[107] identical to columns 1-106, Chinese names
108	fullName	high	[108] identical to columns 1-107, Chinese names
109	fullName	high	[109] identical to columns 1-108, Chinese names
110	fullName	high	[110] identical to columns 1-109, Chinese names
111	fullName	high	[111] identical to columns 1-110, Chinese names
112	fullName	high	[112] identical to columns 1-111, Chinese names
113	fullName	high	[113] identical to columns 1-112, Chinese names
114	fullName	high	[114] identical to columns 1-113, Chinese names
115	fullName	high	[115] identical to columns 1-114, Chinese names
116	fullName	high	[116] identical to columns 1-115, Chinese names
117	fullName	high	[117] identical to columns 1-116, Chinese names
118	fullName	high	[118] identical to columns 1-117, Chinese names
119	fullName	high	[119] identical to columns 1-118, Chinese names
120	fullName	high	[120] identical to columns 1-119, Chinese names
121	fullName	high	[121] identical to columns 1-120, Chinese names
122	fullName	high	[122] identical to columns 1-121, Chinese names
123	fullName	high	[123] identical to columns 1-122, Chinese names
124	fullName	high	[124] identical to columns 1-123, Chinese names
125	fullName	high	[125] identical to columns 1-124, Chinese names
126	fullName	high	[126] identical to columns 1-125, Chinese names
127	fullName	high	[127] identical to columns 1-126, Chinese names
128	fullName	high	[128] identical to columns 1-127, Chinese names
129	fullName	high	[129] identical to columns 1-128, Chinese names
130	fullName	high	[130] identical to columns 1-129, Chinese names
131	fullName	high	[131] identical to columns 1-130, Chinese names
132	fullName	high	[132] identical to columns 1-131, Chinese names
133	fullName	high	[133] identical to columns 1-132, Chinese names
134	fullName	high	[134] identical to columns 1-133, Chinese names
135	fullName	high	[135] identical to columns 1-134, Chinese names
136	fullName	high	[136] identical to columns 1-135, Chinese names
137	fullName	high	[137] identical to columns 1-136, Chinese names
138	fullName	high	[138] identical to columns 1-137, Chinese names
139	fullName	high	[139] identical to columns 1-138, Chinese names
140	fullName	high	[140] identical to columns 1-139, Chinese names
141	fullName	high	[141] identical to columns 1-140, Chinese names
142	fullName	high	[142] identical to columns 1-141, Chinese names
143	fullName	high	[143] identical to columns 1-142, Chinese names
144	fullName	high	[144] identical to columns 1-143, Chinese names
145	fullName	high	[145] identical to columns 1-144, Chinese names
146	fullName	high	[146] identical to columns 1-145, Chinese names
147	fullName	high	[147] identical to columns 1-146, Chinese names
148	fullName	high	[148] identical to columns 1-147, Chinese names
149	fullName	high	[149] identical to columns 1-148, Chinese names
150	fullName	high	[150] identical to columns 1-149, Chinese names
151	fullName	high	[151] identical to columns 1-150, Chinese names
152	fullName	high	[152] identical to columns 1-151, Chinese names
153	fullName	high	[153] identical to columns 1-152, Chinese names
154	fullName	high	[154] identical to columns 1-153, Chinese names
155	fullName	high	[155] identical to columns 1-154, Chinese names
156	fullName	high	[156] identical to columns 1-155, Chinese names
157	fullName	high	[157] identical to columns 1-156, Chinese names
158	fullName	high	[158] identical to columns 1-157, Chinese names
159	fullName	high	[159] identical to columns 1-158, Chinese names
160	fullName	high	[160] identical to columns 1-159, Chinese names
161	fullName	high	[161] identical to columns 1-160, Chinese names
162	fullName	high	[162] identical to columns 1-161, Chinese names
163	fullName	high	[163] identical to columns 1-162, Chinese names
164	fullName	high	[164] identical to columns 1-163, Chinese names
165	fullName	high	[165] identical to columns 1-164, Chinese names
166	fullName	high	[166] identical to columns 1-165, Chinese names
167	fullName	high	[167] identical to columns 1-166, Chinese names
168	fullName	high	[168] identical to columns 1-167, Chinese names
169	fullName	high	[169] identical to columns 1-168, Chinese names
170	fullName	high	[170] identical to columns 1-169, Chinese names
171	fullName	high	[171] identical to columns 1-170, Chinese names
172	fullName	high	[172] identical to columns 1-171, Chinese names
173	fullName	high	[173] identical to columns 1-172, Chinese names
174	fullName	high	[174] identical to columns 1-173, Chinese names
175	fullName	high	[175] identical to columns 1-174, Chinese names
176	fullName	high	[176] identical to columns 1-175, Chinese names
177	fullName	high	[177] identical to columns 1-176, Chinese names
178	fullName	high	[178] identical to columns 1-177, Chinese names
179	fullName	high	[179] identical to columns 1-178, Chinese names
180	fullName	high	[180] identical to columns 1-179, Chinese names
181	fullName	high	[181] identical to columns 1-180, Chinese names
182	fullName	high	[182] identical to columns 1-181, Chinese names
183	fullName	high	[183] identical to columns 1-182, Chinese names
184	fullName	high	[184] identical to columns 1-183, Chinese names
185	fullName	high	[185] identical to columns 1-184, Chinese names
186	fullName	high	[186] identical to columns 1-185, Chinese names
187	fullName	high	[187] identical to columns 1-186, Chinese names
188	fullName	high	[188] identical to columns 1-187, Chinese names
189	fullName	high	[189] identical to columns 1-188, Chinese names
190	fullName	high	[190] identical to columns 1-189, Chinese names
191	fullName	high	[191] identical to columns 1-190, Chinese names
192	fullName	high	[192] identical to columns 1-191, Chinese names
193	fullName	high	[193] identical to columns 1-192, Chinese names
194	fullName	high	[194] identical to columns 1-193, Chinese names
195	fullName	high	[195] identical to columns 1-194, Chinese names
196	fullName	high	[196] identical to columns 1-195, Chinese names
197	fullName	high	[197] identical to columns 1-196, Chinese names
198	fullName	high	[198] identical to columns 1-197, Chinese names
199	fullName	high	[199] identical to columns 1-198, Chinese names
200	fullName	high	[200] identical to columns 1-199, Chinese names

Notes: This dataset appears to be a transposed format where each row represents a hospital facility code (ZZJGDM) and subsequent columns represent patient identifiers and possibly clinical codes. Columns 0-182 appear to be repeated facility names and codes. Columns 183-200 contain SSN-like identifiers and Chinese names. The format is not a typical CSV with a header row; instead, it appears to be a flat table where each row contains a facility code and then a series of patient data spread across many columns.

_______________ods________.csv

1 column1,011 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
1	username	high	[1] header 'A01', values are alphanumeric strings that could be usernames

Notes: Most columns appear to be internal codes and counters. Column 1 contains values that look like usernames (alphanumeric strings), mapped as username. All other columns contain internal IDs, counters, or empty values and are skipped per rules.

_______________ods____csv.csv

18 columns0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
183	lastName	high	[183] value '梁高升' is a Chinese name (last name first)
184	lastName	high	[184] value '卫润七' is a Chinese name (last name first)
185	lastName	high	[185] value '卫润七' is a Chinese name (last name first)
186	lastName	high	[186] value '朱木条姑' is a Chinese name (last name first)
187	lastName	high	[187] value '赵向卓' is a Chinese name (last name first)
188	lastName	high	[188] value '冯霄汉' is a Chinese name (last name first)
189	lastName	high	[189] value '潘敏' is a Chinese name (last name first)
190	lastName	high	[190] value '白真石' is a Chinese name (last name first)
191	lastName	high	[191] value '姚寿卿' is a Chinese name (last name first)
192	lastName	high	[192] value '蒙秀凤' is a Chinese name (last name first)
193	lastName	high	[193] value '申然丰' is a Chinese name (last name first)
194	lastName	high	[194] value '吴桂红' is a Chinese name (last name first)
195	lastName	high	[195] value '秦林苟' is a Chinese name (last name first)
196	lastName	high	[196] value '黎小燕' is a Chinese name (last name first)
197	lastName	high	[197] value '李青山' is a Chinese name (last name first)
198	lastName	high	[198] value '秦田春' is a Chinese name (last name first)
199	lastName	high	[199] value '冷干林' is a Chinese name (last name first)
200	lastName	high	[200] value '舒兰英' is a Chinese name (last name first)

Notes: 201 columns total. Rows 183-200 contain SSN-like identifiers and Chinese names. Other columns appear to be internal codes, counters, or clinical metrics — skipped per rules.

______________hqmsts_2021M1.csv

1 column0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
10	dob	high	[10] header 'A13', values match YYYY-MM-DD date pattern

Notes: Only column 10 (A13) contains valid PII (date of birth). All other columns either contain medical codes/diagnoses, internal IDs, timestamps, or are fully redacted placeholders. No other columns map to defined PII fields based on header names or sample values.

______________hqmsts_2021M2.csv

2 columns0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
10	dob	high	[10] header 'A13', values match YYYY-MM-DD date pattern (e.g., 1941-06-02)
12	country	high	[12] header 'A21C', values are all '中国' (China)

Notes: Only two PII columns identified in first 50 rows: date of birth (A13) and country (A21C). All other columns contain medical codes, timestamps, or redacted values. The dataset appears to be clinical records with extensive ICD codes and quality metrics rather than personal contact information.

______________hqmsts_2021M3.csv

1 column0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
10	dob	high	[10] header 'A13', values match YYYY-MM-DD date pattern (e.g., '1938-08-27', '1997-05-16', '2021-02-26')

Notes: Only column A13 (index 10) contains PII (dates of birth). All other columns either contain internal codes, timestamps, medical codes, or are fully redacted with asterisks/hyphens. No emails, phone numbers, names, addresses, SSNs, passwords, or usernames are present in the first 50 rows.

______________hqmsts_2021M4.csv

1 column0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
10	dob	high	[10] header 'A13', values are dates in YYYY-MM-DD format (1953-09-19, 1955-09-06, 1997-09-30, 2021-03-29). Mixed with what appears to be admission dates, but context is medical records and date pattern matches date-of-birth

Notes: This is a medical records dataset from China's NHC HQMS system. The file contains 226 columns, primarily clinical and hospital quality metrics (diagnosis codes, procedure codes, medical conditions in Chinese, timestamps, numeric metrics). Most columns contain non-PII clinical data, hospital identifiers, codes, or empty placeholder columns. Column [0] 'A01' contains what appear to be internal record IDs (numeric, auto-generated). Column [1] 'A02' contains hospital names (institutional, not personal). Columns [8], [17], [18], [20], [21], [22], [23], [24], [25], [26], [27], [28], [30], [31] contain asterisks/masked values indicating attempted redaction of sensitive data, but the actual values are not visible. Column [10] 'A13' is the only clearly identifiable PII field—it contains dates of birth. All other visible columns contain clinical codes, diagnoses, hospital quality metrics, timestamps, or are empty. The dataset appears to have had personal identifiers redacted (masked with asterisks) but the redaction prevents column type determination for those columns.

______________hqmsts_2021M5.csv

2 columns0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
10	dob	high	[10] header 'A13', values match YYYY-MM-DD date pattern
12	country	high	[12] header 'A21C', values are all '中国' (China)

Notes: 50 rows shown, 226 total columns. Only columns 10 (DOB) and 12 (Country) contain clear PII. All other columns are either internal codes, timestamps, medical codes, or fully redacted values. No emails, phones, names, addresses, SSNs, passwords, usernames, gender, suffix, or facebook IDs are visible in the sample.

______________hqmsts_2021M6.csv

2 columns0 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
10	dob	high	[10] values match YYYY-MM-DD date pattern, header suggests birth date
12	country	high	[12] constant value '中国' (China)

Notes: 226 total columns, only 2 contain PII. Most columns contain medical codes, diagnoses (Cxx), or empty values. Columns 10 (dob) and 12 (country) are the only PII fields identified. All other columns are clinical codes, flags, or empty strings and map to skip per rules.

______________hqmsts______________202104.csv

14 columns7,169 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

____________hqmsts_____20201215172955248820201207.csv

12 columns8,635 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are Chinese full names: 冯科翔, 黄秀, 王林珠
9	gender	high	[9] header 'A12C', values are 1/2 gender codes consistent with male/female coding in Chinese medical records
10	dob	high	[10] header 'A13', values are dates of birth in YYYY-MM-DD format: 1989-05-03, 1978-07-07, 1963-04-06
12	country	high	[12] header 'A15C', values are '中国' (China) — country field
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证): 440882198905030017, 440822197807073127 — the primary government-issued personal identifier
18	address1	high	[18] header 'A22', values are detailed Chinese residential addresses including province/city/district
20	address1	high	[20] header 'A24', values are detailed Chinese addresses — current residential address field
22	address2	medium	[22] header 'A26', values are Chinese addresses — likely a secondary/mailing address field distinct from A24
23	phone	high	[23] header 'A27', values are 11-digit Chinese mobile phone numbers: 18802573860, 15218221213, 13620447689
28	fullName	high	[28] header 'A32', values are Chinese full names: 马玉淦, 黄秀, 王林珠 — likely emergency contact or guardian name
30	address1	medium	[30] header 'A34', values are Chinese addresses — likely emergency contact address
31	phone	high	[31] header 'A35', values are 11-digit Chinese mobile phone numbers: 13750061750, 15218221213, 13620447689 — likely emergency contact phone

Notes: 318-column Chinese medical records (HQMS) file. Cryptic column codes (A11, A12C, A20, etc.) are standard HQMS field identifiers. Key PII fields identified by value inspection: A11=patient full name, A12C=gender code, A13=DOB, A15C=country, A20=Chinese national ID (18-digit, equivalent to SSN), A22/A24=residential addresses, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone. Columns A25C/A28C/A31C appear to be postal/region codes (numeric codes, not zip codes in Western sense — skipped as internal codes). Remaining columns are clinical/medical data: diagnoses (ICD codes), procedures, department codes, financial charges, physician names (B22-B26 are staff names mapped to internal staff IDs in B22C-B26C — skipped as internal), timestamps, and quality metrics.

____________hqmsts_________________HQMS_QQHASDYYY2020.01-09.csv

20 columns29,135 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are Chinese full names (高琴芝, 张景江, 朱凯军, 徐国秀, 任敬生)
9	gender	high	[9] header 'A12C', values 1/2 consistent with gender code (male/female)
10	dob	high	[10] header 'A13', values match YYYY-MM-DD birth date pattern (1961-11-24, 1945-06-12, etc.)
12	country	high	[12] header 'A15C', values are 'CHN' — ISO country code for China
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号)
18	city	high	[18] header 'A22', values are Chinese city/district names (齐齐哈尔市铁锋区, etc.)
20	address1	high	[20] header 'A24', values contain structured Chinese residential addresses with street/village info
22	address2	medium	[22] header 'A26', values contain secondary/current address strings, structured similarly to A24
23	phone	high	[23] header 'A27', values are 11-digit Chinese mobile phone numbers (15845676182, 13763489754, etc.)
28	fullName	high	[28] header 'A32', values are Chinese full names of emergency contacts/relatives (朱德刚, 苗凤兰, etc.)
31	phone	high	[31] header 'A35', values are 11-digit Chinese mobile phone numbers — contact/emergency phone
227	fullName	high	[227] header 'B22', values are Chinese full names (朱艳萍, 张宏伟, 孙玺媛, etc.) — likely attending physician
229	fullName	high	[229] header 'B23', values are Chinese full names (朱艳萍, 张宏伟, etc.) — likely physician role duplicate
231	fullName	high	[231] header 'B24', values are Chinese full names (宋治, 张齐, 姜梅, etc.) — likely physician/staff
233	fullName	high	[233] header 'B25', values are Chinese full names (于航, 杨建明, 陈沫岚, etc.) — likely physician/staff
235	fullName	high	[235] header 'B26', values are Chinese full names (姜丹, 宋飞, 刘思雨, etc.) — likely physician/staff
238	fullName	high	[238] header 'B29', values are Chinese full names (于航, 杨建明, 陈沫岚, etc.) — likely physician/staff
240	fullName	high	[240] header 'B31', values are Chinese full names (宋治, 张齐, 孙玺媛, etc.) — likely physician/staff
241	fullName	high	[241] header 'B32', values are Chinese full names (崔梦玲, 宋飞, 王玮, etc.) — likely physician/staff
251	fullName	medium	[251] header 'C18X01', values are Chinese full names (朱艳萍, 赵辉, 张继东) — likely surgeon/operator name

Notes: 296-column Chinese medical records (HQMS) file. Key PII: col 8 = patient full name, col 10 = DOB, col 17 = 18-digit national ID (equivalent to SSN), cols 23/31 = phone numbers, cols 20/22 = addresses, col 12 = country, col 9 = gender code. Columns 227-241 and 251 contain names of medical staff (physicians, surgeons). Col 28 is likely next-of-kin/emergency contact name. The bulk of remaining columns are clinical/diagnostic codes, ICD codes, procedure codes, timestamps, and hospital quality metrics — all skipped.

__________hqmsts_M202001.csv

14 columns3,350 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

__________hqmsts_M202002.csv

14 columns2,320 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

__________hqmsts_M202003.csv

14 columns3,025 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

__________hqmsts_M202004.csv

14 columns3,232 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

__________hqmsts_M202005.csv

14 columns3,324 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

__________hqmsts_M202006.csv

14 columns3,346 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

__________hqmsts_M202007.csv

14 columns3,551 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

__________hqmsts_M202008.csv

14 columns3,609 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

__________hqmsts_M202009.csv

14 columns3,465 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

__________hqmsts_M202010.csv

14 columns3,531 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

_________hqmsts2020______________20210125101154.csv

8 columns7,942 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] A11: garbled-encoded Chinese patient names in medical record context
9	gender	high	[9] A12C: values 1/2 are gender codes (male/female) in Chinese HQMS medical records
10	dob	high	[10] A13: datetime values ranging 1949–2020, consistent with patient date of birth
17	ssn	high	[17] A20: 18-digit Chinese resident identity card numbers (居民身份证), national ID equivalent to SSN
20	address1	medium	[20] A24: encoded Chinese text of varying length consistent with patient current address
22	address2	medium	[22] A26: encoded Chinese text consistent with patient registered/household address, parallel to A24
28	fullName	medium	[28] A32: short encoded Chinese strings consistent with contact/next-of-kin names
31	phone	high	[31] A35: 11-digit Chinese mobile phone numbers (e.g. 18113224088, 13378370082)

Notes: 290-column HQMS medical records file. Columns use opaque field codes (A11, A12C, etc.). Most columns are clinical/diagnostic data (ICD codes, procedure codes, admission timestamps, ward codes). PII is limited to: patient name [8], gender code [9], DOB [10], national ID [17], addresses [20][22], contact name [28], phone [31]. Text encoding is corrupted (likely GBK/GB2312 read as Latin-1), producing garbled characters throughout Chinese-language fields.

_________hqmsts_M20210301-20210330_1.csv

12 columns7,429 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are Chinese full names (罗耀胜, 彭海贞, 谢旭, 杨英全)
9	gender	high	[9] header 'A12C', values are 1/2 — standard Chinese medical record gender codes (1=male, 2=female)
10	dob	high	[10] header 'A13', values are YYYY-MM-DD birth dates (1941-11-18, 1964-03-27, etc.)
12	country	high	[12] header 'A15C', values are '中国' (China) — nationality/country field
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) — e.g. 440202194111180950
18	address1	high	[18] header 'A22', values are Chinese residential addresses (广东省清远市清城区...)
20	address2	high	[20] header 'A24', values are detailed street/unit addresses with postal prefix (441802000广东省清远市清城区@南岸一街8号...)
22	address1	medium	[22] header 'A26', values are alternate/contact addresses similar in format to A24 — likely a second address field
23	phone	high	[23] header 'A27', values are 11-digit Chinese mobile numbers (13802890060, 13435283982, 13602946741)
28	fullName	high	[28] header 'A32', values are Chinese full names (罗文伟, 吴志辉, 邓彩红, 赖秋连) — likely emergency contact / next of kin
30	address1	medium	[30] header 'A34', values are Chinese addresses — likely emergency contact address
31	phone	high	[31] header 'A35', values are 11-digit Chinese mobile numbers (13425201299, 13802890060, 15875915130) — likely emergency contact phone

Notes: 226-column Chinese medical records file (HQMS format). PII is concentrated in columns 8–31 (patient demographics block). Column 17 (A20) contains 18-digit Chinese national ID numbers which encode DOB and gender and function as the national SSN equivalent. Columns 28/30/31 appear to be next-of-kin/emergency contact name, address, and phone. The vast majority of remaining columns (C06–F09 series) are ICD-10 diagnosis codes, procedure codes, and clinical quality metrics — all non-PII.

_________hqmsts_M20210401-20210430_1.csv

12 columns7,664 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are Chinese personal names (罗耀胜, 叶青, 黎丽霞, 刘汝容, 梁焕容)
9	gender	high	[9] header 'A12C', values are 1/2 — standard gender code (1=male, 2=female) in Chinese medical records
10	dob	high	[10] header 'A13', values are YYYY-MM-DD birth dates (1941-11-18, 1929-03-27, 1963-02-04)
12	country	high	[12] header 'A15C', values are '中国' (China) — country of nationality field in Chinese medical records
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (440202194111180950, 440127192903270017) — China's resident identity card number
18	address1	high	[18] header 'A22', values are Chinese residential addresses (广东省清远市清城区...)
20	address2	medium	[20] header 'A24', values appear to be a second address field with area code prefix and '@' separator (441802000广东省清远市清城区@南岸一街8号之二一梯301)
22	address1	medium	[22] header 'A26', values are Chinese addresses similar to A22/A24 — likely current/registered address variant
23	phone	high	[23] header 'A27', values are 11-digit Chinese mobile phone numbers (13425201299, 13376676567, 13802892816)
28	fullName	high	[28] header 'A32', values are Chinese personal names (郑淑华, 叶剑明, 卢就华) — likely next-of-kin/emergency contact name
30	address1	medium	[30] header 'A34', values are Chinese addresses — likely emergency contact address
31	phone	high	[31] header 'A35', values are 11-digit Chinese mobile phone numbers (13425201299, 13376676567, 13802892816) — likely emergency contact phone

Notes: 226-column Chinese medical record (HQMS) file. Key PII: patient name (A11), gender code (A12C), DOB (A13), nationality (A15C), 18-digit national ID/SSN (A20), residential addresses (A22, A24, A26), mobile phone (A27), next-of-kin name (A32), next-of-kin address (A34), next-of-kin phone (A35). Columns C06x*–C08x* contain ICD diagnosis codes and Chinese disease names (clinical data, not PII). Remaining columns are hospital identifiers, admission/discharge timestamps, department codes, and clinical quality metrics — all skipped.

_________hqmsts_M20210501-20210531_1.CSV

12 columns7,239 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are Chinese personal names (罗耀胜, 成金旺, 黄会香, 严颂伟, 钟惠珍)
9	gender	high	[9] header 'A12C', values are 1/2 — standard Chinese medical record gender codes (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO date patterns matching birthdates (1941-11-18, 1939-11-23, 1948-08-16)
12	country	high	[12] header 'A15C', values are '中国' (China) — country of origin field
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) — e.g. 440202194111180950
18	address1	high	[18] header 'A22', values are Chinese residential addresses (广东省清远市清城区凤城街道...)
20	address1	high	[20] header 'A24', values are detailed mailing addresses with street-level info (广东省清远市清城区凤城街道@南岸一街8号...)
22	address2	medium	[22] header 'A26', values are similar addresses to A24 but appear to be a secondary/contact address field
23	phone	high	[23] header 'A27', values are 11-digit Chinese mobile numbers (13417221135, 13750176280, 13828575680)
28	fullName	high	[28] header 'A32', values are Chinese personal names (郑淑华, 成锐波, 余新铝, 严莹, 钟惠玲) — likely next-of-kin/emergency contact name
30	address1	high	[30] header 'A34', values are Chinese residential addresses — likely contact/emergency address
31	phone	high	[31] header 'A35', values are 11-digit Chinese mobile numbers (13425201299, 13750176280, 13828507568) — likely contact/emergency phone

Notes: 226-column Chinese medical records file (HQMS format). PII is concentrated in columns 8–31 (patient demographics block). A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit national ID (mapped as ssn), A22/A24=home addresses, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone. Remaining columns are clinical/diagnostic codes (ICD codes, diagnosis names, procedure flags), hospital identifiers, admission/discharge timestamps, and department codes — all non-PII.

_________hqmsts_M20210601-20210630_1.CSV

12 columns7,297 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are Chinese personal names (江金带, 林凯华, 祝远彩, 林水带, 谭瑞广)
9	gender	high	[9] header 'A12C', values are 1/2 which in Chinese medical records encode gender (1=male, 2=female)
10	dob	high	[10] header 'A13', values are dates in YYYY-MM-DD format (1940-09-02, 1950-08-15) consistent with patient date of birth
12	country	high	[12] header 'A15C', values are '中国' (China) — country of origin/nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (440127194009021445, 440202195008150328) — China's national identity card number (身份证号)
18	address1	high	[18] header 'A22', values are Chinese residential addresses (广东省清远市清城区...)
20	address1	high	[20] header 'A24', values are detailed Chinese addresses with street/village information — current address field
22	address1	high	[22] header 'A26', values are Chinese addresses matching A24 pattern — registered/household address field
23	phone	high	[23] header 'A27', values are 11-digit Chinese mobile phone numbers (13326548114, 13631080693, 18948041278)
28	fullName	high	[28] header 'A32', values are Chinese personal names (洪国安, 周燕, 程柏卢, 曾庆健, 冯秀桃) — likely next-of-kin/emergency contact name
30	address1	high	[30] header 'A34', values are Chinese addresses — contact/emergency contact address
31	phone	high	[31] header 'A35', values are 11-digit Chinese mobile phone numbers (13828575866, 13828500541) — emergency contact phone

Notes: 226-column Chinese medical records file from HQMS system. Key PII fields identified by value inspection since headers are opaque HQMS codes. Column 17 (A20) contains Chinese 18-digit national ID numbers (身份证号) which encode DOB and gender but are mapped as ssn as they are the primary national identifier. Columns 8 and 28 contain patient name and emergency contact name respectively. Columns 20/22/30 contain overlapping address fields (current, registered, and contact addresses). Columns 23 and 31 contain patient and emergency contact phone numbers. Remaining columns contain medical/clinical data: ICD codes, diagnoses, admission/discharge dates, department codes, hospital identifiers, and quality metrics — all skipped.

________hqmsts_M20200924_1.csv

14 columns3,104 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are garbled-encoding Chinese patient names (e.g., 'Īΰ' is a 2-char Chinese name)
9	gender	high	[9] header 'A12C', values '1'/'2' consistent with Chinese HQMS gender code (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like '1951-10-05', '1954-06-03' — patient birth dates
12	country	high	[12] header 'A15C', garbled-encoding values consistently represent '中国' (China) — patient nationality
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (居民身份证号) e.g., '440921195110054233' — the primary national identifier
18	address1	high	[18] header 'A22', garbled-encoding values are Chinese province/city strings — patient registered address (户籍地址)
20	address1	high	[20] header 'A24', garbled-encoding values are detailed street-level addresses e.g., containing road names and house numbers — patient current address
21	zip	high	[21] header 'A25C', values like '525327', '525300' are 6-digit Chinese postal codes for the patient's address
23	phone	high	[23] header 'A27', values include '13432316787', '13824874123', '13692654546' — 11-digit Chinese mobile phone numbers
28	fullName	high	[28] header 'A32', garbled-encoding Chinese names (emergency contact or next-of-kin name)
30	address1	high	[30] header 'A34', garbled-encoding detailed address strings matching same pattern as A24 — contact/next-of-kin address
31	phone	high	[31] header 'A35', values '15919274255', '13926710791', '18802517532' — 11-digit Chinese mobile numbers (emergency contact phone)
240	fullName	high	[240] header 'B31', garbled-encoding Chinese names consistent with patient/surgeon name field
241	fullName	high	[241] header 'B32', garbled-encoding Chinese names — likely secondary clinician/surgeon name

Notes: 296-column HQMS medical records CSV with garbled encoding (likely GBK/GB2312 read as Latin-1). Column headers are HQMS field codes (A01, A11, etc.). Key PII: A11=patient name, A12C=gender code, A13=DOB, A15C=nationality, A20=18-digit Chinese national ID (SSN equivalent), A22=registered address, A24=current address, A25C=postal code, A27=patient phone, A32=emergency contact name, A34=emergency contact address, A35=emergency contact phone, B31/B32=clinician names. Majority of columns are clinical/diagnostic codes (ICD codes in C06x* series), procedure codes, timestamps, and internal flags — all skipped.

________hqmsts_____20201204162756448510_.csv

22 columns2,727 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are Chinese personal names (马成强, 廖宝林, 魏生兰) — patient name field
9	gender	medium	[9] header 'A12C', binary values 1/2 consistent with gender code in Chinese medical records (1=male, 2=female)
10	dob	high	[10] header 'A13', values are ISO dates like 1957-06-10, 1963-10-18 — patient date of birth
12	country	high	[12] header 'A15C', values are '中国' (China) — nationality/country field
17	ssn	high	[17] header 'A20', 18-digit Chinese national ID numbers (居民身份证号) — national identifier equivalent to SSN
18	address1	high	[18] header 'A22', values are Chinese address strings including province/city/district — primary address
20	address1	high	[20] header 'A24', values are detailed Chinese street addresses — residential address line
21	zip	high	[21] header 'A25C', values like '839000' are Chinese postal codes
22	address2	medium	[22] header 'A26', values are address strings that appear to be a second/contact address field
23	phone	high	[23] header 'A27', 11-digit numbers matching Chinese mobile phone format (138/135/099 prefixes)
24	zip	high	[24] header 'A28C', values like '839000' are Chinese postal codes for work/contact address
26	phone	high	[26] header 'A30', 11-digit Chinese phone numbers — second contact phone field
28	fullName	high	[28] header 'A32', values are Chinese personal names (马冠茹, 刘淑文) — emergency contact name
30	address1	high	[30] header 'A34', values are Chinese address strings — emergency contact address
31	phone	high	[31] header 'A35', 11-digit Chinese phone numbers — emergency contact phone
227	fullName	high	[227] header 'B22', values are Chinese personal names (孙红霞, 多尔坤·沙衣热木) — likely attending physician name
229	fullName	high	[229] header 'B23', values are Chinese personal names — secondary physician/staff name
231	fullName	high	[231] header 'B24', values are Chinese personal names — tertiary physician/staff name
233	fullName	high	[233] header 'B25', values are Chinese personal names — additional staff name field
235	fullName	high	[235] header 'B26', values are Chinese personal names — additional staff name field
238	fullName	high	[238] header 'B29', values are Chinese personal names (李莎, 毛苏伟) — staff/reviewer name field
240	fullName	high	[240] header 'B31', values are Chinese personal names — additional medical staff name field

Notes: 296-column Chinese HQMS medical records file. Column A20 (idx 17) contains 18-digit Chinese national ID numbers mapped as SSN. Multiple fullName fields represent patient (A11), emergency contact (A32), and various medical staff roles (B22–B31). Columns A12C (gender code 1/2) and A15C (country) are standard coded fields. Most remaining columns are ICD diagnosis codes, procedure codes, clinical metrics, and internal hospital reporting flags — all skipped.

________hqmsts_____2020120416341761239_.csv

19 columns2,758 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are Chinese personal names (patient names)
9	gender	medium	[9] header 'A12C', values '1'/'2' consistent with gender code field in Chinese medical records
10	dob	high	[10] header 'A13', values are ISO dates like '2019-11-22', '1952-08-14' — date of birth
12	country	high	[12] header 'A15C', values '中国' (China) — country of patient
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (身份证号) — national identity number equivalent to SSN
20	address1	high	[20] header 'A24', values are detailed Chinese street addresses — primary address
21	zip	high	[21] header 'A25C', values '839000' — Chinese postal codes
22	address2	high	[22] header 'A26', values are Chinese addresses (secondary/mailing address field)
23	phone	high	[23] header 'A27', values are 11-digit Chinese mobile phone numbers
26	phone	high	[26] header 'A30', values are 11-digit Chinese mobile phone numbers (emergency contact phone)
28	fullName	high	[28] header 'A32', values are Chinese personal names (emergency contact names)
31	phone	high	[31] header 'A35', values are 11-digit Chinese mobile phone numbers (contact phone)
227	fullName	high	[227] header 'B22', values are Chinese personal names (medical staff)
229	fullName	high	[229] header 'B23', values are Chinese personal names (medical staff)
231	fullName	high	[231] header 'B24', values are Chinese personal names (medical staff)
233	fullName	high	[233] header 'B25', values are Chinese personal names (medical staff)
235	fullName	high	[235] header 'B26', values are Chinese personal names (medical staff)
238	fullName	high	[238] header 'B29', values are Chinese personal names (medical staff)
240	fullName	high	[240] header 'B31', values are Chinese personal names (medical staff)

Notes: 296-column Chinese HQMS medical records file. Headers use internal Chinese medical reporting codes (A11, A12C, etc.). PII identified via sample values: patient name, DOB, gender, national ID (18-digit 身份证, mapped as SSN), address, postal code, phone numbers for patient and emergency contacts, and names/phones of multiple medical staff members. Remaining columns are clinical/diagnostic data (ICD codes, diagnoses, procedure codes, timestamps, hospital identifiers) — all skipped.

________hqmsts_____202012091830044839202011.csv

25 columns2,736 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] header 'A11', values are Chinese personal names (孙守江, 卓梅英, 陈瑶, 杨志刚, 马桂花)
9	gender	high	[9] header 'A12C', values are 1/2 consistent with gender code (male/female)
10	dob	high	[10] header 'A13', values are dates of birth in YYYY-MM-DD format (1937-03-10, 1948-05-25, etc.)
12	country	high	[12] header 'A15C', values are '中国' (China) — country field
17	ssn	high	[17] header 'A20', values are 18-digit Chinese national ID numbers (resident identity card numbers)
18	city	high	[18] header 'A22', values are Chinese city/region names (新疆哈密, 江苏省徐州市)
20	address1	high	[20] header 'A24', values are detailed street addresses in Chinese
21	zip	high	[21] header 'A25C', values are 6-digit Chinese postal codes (839000, 839200)
22	address2	high	[22] header 'A26', values are detailed addresses, likely current/mailing address variant
23	phone	high	[23] header 'A27', values are 11-digit Chinese mobile phone numbers (13319028575, 13201060635)
25	address1	medium	[25] header 'A29', values are addresses (workplace/next-of-kin address)
26	phone	high	[26] header 'A30', values are 11-digit Chinese mobile phone numbers — contact/emergency phone
28	fullName	high	[28] header 'A32', values are Chinese personal names (孙东玲, 郭同步, 郜媛) — next-of-kin or contact name
30	address1	medium	[30] header 'A34', values are Chinese addresses — next-of-kin/contact address
31	phone	high	[31] header 'A35', values are 11-digit Chinese mobile phone numbers — next-of-kin/contact phone
214	fullName	high	[214] header 'B22', values are Chinese personal names (多尔坤·沙衣热木, 张积利, 杨荣) — medical staff names
216	fullName	high	[216] header 'B23', values are Chinese personal names — additional medical staff
218	fullName	high	[218] header 'B24', values are Chinese personal names (王磊, 阿迪力·牙生, 杨荣)
220	fullName	high	[220] header 'B25', values are Chinese personal names (魏超豹, 阿迪力·牙生, 邹贝)
222	fullName	high	[222] header 'B26', values are Chinese personal names (马福莉, 郭丹, 金苗, 帕提古丽)
225	fullName	high	[225] header 'B29', values are Chinese personal names (毛苏伟, 李莎) — likely attending physician
227	fullName	high	[227] header 'B31', values are Chinese personal names (多尔坤·沙衣热木, 张积利, 杨荣, 冯轶荣)
238	fullName	high	[238] header 'C18X01', values are Chinese personal names (魏超豹, 冯轶荣) — surgical/procedure staff names
239	fullName	high	[239] header 'C19x01', values are Chinese personal names (李剑春) — assistant surgeon names
244	fullName	high	[244] header 'C23X01', values are Chinese personal names (魏超豹, 冯轶荣) — anesthesiologist or other staff names

Notes: 293 columns total; medical records from China's HQMS system. Key PII: patient name (A11), gender code (A12C), DOB (A13), nationality (A15C), national ID/SSN (A20 — 18-digit Chinese resident ID), city (A22), address (A24/A26/A29/A34), postal codes (A25C), phone numbers (A27/A30/A35), next-of-kin name (A32) and address (A34). Multiple medical staff full names appear in B22/B23/B24/B25/B26/B29/B31/C18X01/C19x01/C23X01. Remaining columns are clinical/diagnostic codes, timestamps, ICD codes, procedure codes, and internal identifiers — all skipped.

_______hqmsts_202009_202009.csv

11 columns1,751 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	medium	[8] header 'A11', values are short Chinese text consistent with patient names (encoding garbled)
9	gender	medium	[9] header 'A12C', values are 1/2 consistent with gender code fields in Chinese medical systems
10	dob	high	[10] header 'A13', values are ISO date strings like '1954-03-19' — patient date of birth
12	country	medium	[12] header 'A15C', garbled Chinese text consistent with nationality/country field in HQMS schema
17	ssn	high	[17] header 'A20', 18-digit values like '130321195403198710' are Chinese Resident Identity Card numbers — national ID equivalent to SSN
18	address1	medium	[18] header 'A22', long garbled Chinese strings consistent with patient home address
23	phone	high	[23] header 'A27', 11-digit values like '13473844580' are Chinese mobile phone numbers
26	phone	high	[26] header 'A30', same 11-digit Chinese mobile numbers — likely emergency contact or alternate phone
28	fullName	medium	[28] header 'A32', short garbled Chinese text consistent with contact/next-of-kin person name
30	address1	medium	[30] header 'A34', long garbled Chinese strings consistent with contact/next-of-kin address
31	phone	high	[31] header 'A35', 11-digit Chinese mobile numbers matching A27/A30 pattern — contact phone number

Notes: 296-column Chinese HQMS medical records file with encoding issues (mojibake). Column headers follow HQMS standard field codes (A11=patient name, A12C=gender, A13=DOB, A15C=nationality, A20=national ID card number, A22=home address, A27/A30/A35=phone numbers, A32=contact name, A34=contact address). Remaining columns are clinical/diagnostic data: ICD codes, procedure codes, diagnosis names, admission/discharge dates, ward codes, surgical records — all non-PII medical metadata.

_______hqmsts_202010_202010.csv

16 columns1,706 rows

File structure

Format: CSV·Delimiter: Comma·Has header: yes·Quote: "

Source column	Mapped field	Confidence	LLM assessment
8	fullName	high	[8] A11 — garbled Chinese character values consistent with patient names in Chinese medical records
9	gender	high	[9] A12C — values 1/2 consistent with Chinese medical gender codes (1=male, 2=female)
10	dob	high	[10] A13 — values in YYYY-MM-DD format, e.g. 1964-09-27, clearly dates of birth
12	country	medium	[12] A15C — garbled values consistent with 中国 (China), nationality/country field
17	ssn	high	[17] A20 — 18-digit Chinese national ID numbers (居民身份证号), encodes DOB and gender
18	address1	high	[18] A22 — garbled Chinese address strings, patient registered address
20	address2	medium	[20] A24 — same address pattern as A22, likely current/mailing address variant
21	zip	medium	[21] A25C — 6-digit codes (e.g. 066599) consistent with Chinese postal codes for registered address
22	address1	medium	[22] A26 — third address field with same Chinese address pattern, contact address
23	phone	high	[23] A27 — 11-digit values (e.g. 18330369781) matching Chinese mobile phone number format
24	zip	medium	[24] A28C — 6-digit postal codes associated with contact address
26	phone	high	[26] A30 — 11-digit Chinese mobile numbers, second phone field (same values as A27 in sample)
27	zip	medium	[27] A31C — 6-digit postal codes, third zip associated with emergency contact address
28	fullName	high	[28] A32 — short Chinese text values consistent with emergency contact/next-of-kin names
30	address1	medium	[30] A34 — short Chinese address text, emergency contact address
31	phone	high	[31] A35 — 11-digit Chinese mobile numbers, emergency contact phone

Notes: 296-column Chinese medical records CSV from HQMS system. Column headers use internal HQMS codes (A11, A12C, etc.) rather than descriptive names. Columns identified by value patterns: 18-digit national ID in A20 mapped as SSN, DOB in A13, gender code in A12C, multiple address+zip pairs (A22/A25C, A24/A28C, A26/A31C), three phone fields (A27, A30, A35), two name fields (A11=patient, A32=emergency contact). Remaining 280+ columns are clinical/diagnostic data (ICD codes, procedure codes, admission/discharge dates, lab flags) and are skipped.

Data found in this dataset

Search this dataset

Source files

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure

File structure