Blog: Keeping Medical Data Private

Keeping Medical Data Private
Technology Review (04/13/10) Gammon, Katharine

Vanderbilt University (VU) researchers have developed an algorithm designed to protect the privacy of medical patients while maintaining researchers' ability to analyze large amounts of genetic and clinical data. Although patient records are anonymized, they still contain the numerical codes, known as ICD codes, which represent every condition a doctor has detected. As a result, VU professor Bradley Malin says it is possible to follow a specific set of codes backward and identify a person. Malin and his colleagues found that they could identify more than 96 percent of a group of patients based only on their particular set of ICD codes. To make patients more private, the researchers designed an algorithm that searches a database for combinations of ICD codes that distinguish a patient and then substitutes a more general version of the codes to ensure each patient's altered record is indistinguishable from a certain number of other patients. The researchers tested the algorithm on 2,762 patients and could not identify any of them based on their new ICD codes.

View Full Article