Meddies-PII: A Multilingual Framework for Personally Identifiable Information Extraction in Clinical De-identification
This work introduces Meddies-PII-Dataset, a corpus of one million synthetic clinical documents spanning seventeen languages and nine PII labels, and trains Meddies-PII-Model, a BIOES token classifier, and compares it with existing PII extraction systems using exact-match entity-level F1.
Linh Le, Ho Dinh Tri, Christian Hoang et al.
· 0 citations