Health data privacy: A case of undesired inferences

摘要

In this work, we investigate privacy violations that occur when non-confidential medical data is combined with domain ontologies to infer confidential data. We propose a framework to detect such privacy violations and to eliminate undesired inferences. Our inference channel removal is based on modifying data that contribute to an inference. We show that our method is sound and complete. Soundness means that we modify only data items that lead to undesired inferences. Completeness means that we detect all inferences leading to undesired data disclosures. Finally, we show that our approach preserves data availability by minimizing the number of data items to be modified. An important aspect of our approach is that it sets the foundation for creating patient-specific privacy policies; an emerging need in the healthcare domain.