Recognition and pseudonymisation of medical records for secondary use

Health records rank among the most sensitive personal information existing today. An unwanted disclosure to unauthorised parties usually results in significant negative consequences for an individual. Therefore, health records must be adequately protected in order to ensure the individual’s privacy. However, health records are also valuable resources for clinical studies and research activities. In order to make the records available for privacy-preserving secondary use, thorough de-personalisation is a crucial prerequisite to prevent re-identification. This paper introduces MEDSEC, a system which automatically converts paper-based health records into de-personalised and pseudonymised documents which can be accessed by secondary users without compromising the patients’ privacy. The system converts the paper-based records into a standardised structure that facilitates automated processing and the search for useful information.