Abstract Anonymization of General Practioner Medical Records

The Electronic Patient Record (EPR) is both a legal document and a tool for use by physicians and other health personnel during provision of health care. Its primary purpose is to provide and store information about the patient in clinical settings, but it’s also a source of medical knowledge (e.g. epidemiology and quality of care). Due to the sensitive nature of the data they must be handled in a secure manner with a high awareness of privacy concerns. This problem can be partially avoided by applying an anonymization procedure to the data. For large volumes of data (e.g. thousands of patient records) such a procedure must be partially automated. We aim to develop techniques and methods for semi-automated anonymization of medical record information. We first present the requirements and goals of anonymization. Relevant goals for designing an anonymization method are: complying with national laws, and making the anonymization as automated as possible. We discuss anonymization challenges, including linguistic issues (e.g. spelling and ambiguity) and determining which parts of the data that is sensitive. Finally we propose methods including utilization of database structure, dictionaries, heuristics and natural language processing for anonymizing patient records in general, but with focus on general practioner records gathered from a Profdoc Vision database