Querying Medical Datasets while Preserving Privacy

This paper addresses the challenge of identifying clinically-relevant patterns in medical datasets without endangering patient privacy. To this end, we treat medical datasets as black box for both internal and external users of the data enabling a remote query mechanism to construct and execute database queries. The novelty of the approach lies in avoiding the complex data de-identification process which is often used to preserve patient privacy. The implemented toolkit combines software engineering technologies such as Java EE and RESTful web services, to allow exchanging medical data in an unidentifiable XML format along with restricting users to the need-to-know privacy principle. Consequently, the technique inhibits retrospective processing of data, such as attacks by an adversary on a medical dataset using advanced computational methods to reveal Protected Health Information (PHI). The approach is validated on an endoscopic reporting application based on openEHR and MST standards. The proposed approach is largely motivated by the issues related to querying datasets by clinical researchers, governmental or non-governmental organizations in monitoring health care services to improve quality of care.