figshare
Browse
12911_2024_2422_MOESM1_ESM.docx (16.05 kB)

Additional file 1 of Automatic de-identification of French electronic health records: a cost-effective approach exploiting distant supervision and deep learning models

Download (16.05 kB)
journal contribution
posted on 2024-02-17, 04:40 authored by Mohamed El Azzouzi, Gouenou Coatrieux, Reda Bellafqira, Denis Delamarre, Christine Riou, Naima Oubenali, Sandie Cabon, Marc Cuggia, Guillaume Bouzillé
Additional file 1. The details of the distribution of document types in the sample. The distribution of document types within the sample is provided, with document categories and corresponding codes. The most common document types include various questionnaires related to medical specialties, such as anesthesia, prescription, physiotherapy, as well as different types of medical records. In addition, there are documents labeled as "Full Document" and a variety of other types of medical questionnaires, each with varying frequency in the dataset.

History