deid-clinical-notes

name	deid-clinical-notes
description	Detect and label de-identification–target personal information (names, IDs, addresses, dates/times/periods, other hospitals) in unstructured Korean clinical notes using regular-expression rules, wrapping each match as <LABEL>string</LABEL>. Sensitive-diagnosis rules (KCD/name) ship but are off by default. Use for LYDUS pseudonymization/de-identification of free-text medical records.
tier	community
category	lydus
parameters	{"text":{"description":"Free text (or path to a .txt file) to label. Alternatively use csv+column for a CSV column.","type":"string"},"configs":{"description":"List of YAML config files defining the regex rules. Default = disease_kcd + disease_name + smc.","type":"array"}}

De-identification Labeling for Korean Clinical Notes

Automatically detects de-identification–target information in unstructured Korean free-text medical records using regular-expression rules, and wraps each detected span as <LABEL>string</LABEL>. Based on the MOHW Medical Data Utilization Guidelines and Samsung Medical Center internal policy; the regex rules were derived from how these identifiers actually appear in real records.

input : 24.08.01 퇴원 후 hemoptysis 발생
output: <DATE>24.08.01</DATE> 퇴원 후 hemoptysis 발생

When to Use This Skill