MEDDOCAN

MEDDOCAN: Medical Document Anonymization Track This dataset is designed for the MEDDOCAN task, sponsored by Plan de Impulso de las Tecnologías del Lenguaje. It is a manually classified collection of 1,000 clinical case reports derived from the Spanish Clinical Case Corpus (SPACCC), enriched with PHI expressions. The annotation of the entire set of entity mentions was carried out by experts annotatorsand it includes 29 entity types relevant for the annonymiation of medical documents.22 of these annotation types are actually present in the corpus: TERRITORIO, FECHAS, EDAD_SUJETO_ASISTENCIA, NOMBRE_SUJETO_ASISTENCIA, NOMBRE_PERSONAL_SANITARIO, SEXO_SUJETO_ASISTENCIA, CALLE, PAIS, ID_SUJETO_ASISTENCIA, CORREO, ID_TITULACION_PERSONAL_SANITARIO,ID_ASEGURAMIENTO, HOSPITAL, FAMILIARES_SUJETO_ASISTENCIA, INSTITUCION, ID_CONTACTO ASISTENCIAL,NUMERO_TELEFONO, PROFESION, NUMERO_FAX, OTROS_SUJETO_ASISTENCIA, CENTRO_SALUD, ID_EMPLEO_PERSONAL_SANITARIO For further information, please visit temu.bsc.es/meddocan/ or send an email to encargo-pln-life@bsc.es