Recognition and pseudonymisation of medical records for secondary use
- 212 Downloads
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.
KeywordsDe-personalisation Information management Secondary use Privacy Pseudonymisation
We thank our business partners XiTrust Secure Technologies and Xylem Technologies for supporting the implementation of the case studies carried out within the MEDSEC project. The research was funded by BRIDGE (#824884), FFG-Austrian Research Promotion Agency, and supported by COMET K1, FFG-Austrian Research Promotion Agency.
Since real-life records from a hospital archive with personal data were used in the case study, special care was taken to ensure the involved patients’ privacy. Access to the data was only allowed for the directly involved project members. Furthermore, the test data were only accessible within the archive computer network and records were not stored, copied, or processed outside the network environment.
- 2.Appelt DE (1999) Introduction to information extraction. AI Commun 12(3):161–172Google Scholar
- 6.Galindo D, Verheul ER (2007) Microdata sharing via pseudonymisation. Joint UNECE/Eurostat Work Session on Statistical Data Confidentiality, Manchester, pp 24–32Google Scholar
- 9.Health Level Seven International (2007) HL7 version 3. Online: www.hl7.org
- 11.Heurix J, Rella A, Fenz S, Neubauer T (2013) Automated transformation of semi-structured text elements. In: Proceedings of America’s conference on information systems (AMCIS), pp 1–11Google Scholar
- 18.Sibanda T, He T, Szolovits P, Uzuner O (2006) Syntactically-informed semantic category recognition in discharge summaries. In: AMIA annual symposium proceedings, pp 714–718Google Scholar
- 21.Union European (1995) Directive 95/46/EC of the European Parliament and of the Council of 24 October 1995 on the protection of individuals with regard to the processing of personal data and on the free movement of such data. Off J Eur Commun L281:31–50Google Scholar
- 22.United States Congress (1996) Health insurance portability and accountability Act of 1996. Pub.L. 104–191, 110 Stat. 1936Google Scholar