Research project Health Bank de-identification tool and its practical use
The aim of this project is to enable the practical use of methods for de-identification of clinical free text written in Swedish. The HB Deid tool will be tested on textual patient data from Region Stockholm and not only in a laboratory environment at Stockholm University.
The aim of this project is to develop methods for de-identification of clinical free text, written in Swedish, that can be put to practical use.
The identity of patients will not be revealed in the process, the method enables disclosure control. The result is data describing symptoms, diagnoses and outcomes of treatments that may be used safely for medical research, for example around COVID-19.
The HB Deid de-identification tool has been developed in the academy and will be installed at Region Stockholm for pilot studies. New requirements on the tool will be collected and implemented in future versions of HB Deid.
HB Deid is a tool that has been developed for de-identification of texts. See how it works
Project members
Project managers
Hercules Dalianis
Professor
Publications
Vakili, T., Lamproudis, A., Henriksson, A. and Dalianis, H. (2022)
“Downstream Task Performance of BERT Models Pre-Trained Using Automatically De-Identified Clinical Data” (submitted)