Local-first multilingual clinical text de-identification framework with Dutch and English models, pluggable language profiles, and Python, CLI, batch, and HTTP interfaces.
-
Updated
Sep 21, 2026 - Python
Local-first multilingual clinical text de-identification framework with Dutch and English models, pluggable language profiles, and Python, CLI, batch, and HTTP interfaces.
Data import, splitting, and synthetic generation for MedDeID.
Legacy redirect to the canonical MedDeID website and documentation.
Gold core-PII character subannotation for MedDeID.
Dutch nl-BE and nl-NL language profiles and resources for MedDeID.
Canonical MedDeID schema, taxonomy, normalization, and validation.
Optional multi-reviewer reconciliation for MedDeID.
Primary-span annotation application for MedDeID.
Model training, selection, refit, and bundle export for MedDeID.
To associate your repository with the meddeid topic, visit your repo's landing page and select "manage topics."