"A Cryptography Game-Changer for Biomedical Research at Scale"

The P4 approach, which encompasses predictive, personalized, participatory, precision, and preventive medicine, is the future of healthcare. In order to increase its adoption and effectiveness, it is essential for clinical data on large numbers of individuals to be shared efficiently among all shareholders. However, gathering data is a challenge as it is isolated between individual hospitals, medical practices, and clinics globally. There are also privacy risks that come from the disclosure of medical data. Without effective privacy-preserving technologies, these risks present a barrier to enhancing P4 medicine. Existing methods either provide only limited protection for patients' privacy by requiring institutions to share intermediate results or do away with the accuracy of results by adding noise to the data to prevent possible data leakage. Therefore, researchers at EPFL's (Ecole Polytechnique Fédérale de Lausanne) Laboratory for Data Security, in collaboration with colleagues at Lausanne University Hospital, MIT's Computer Science and Artificial Intelligence Laboratory (CSAIL), and the Broad Institute of MIT and Harvard, developed a novel federated analytics system called FAMHE. This system allows different healthcare providers to collaboratively conduct statistical analyses and develop Machine Learning (ML) models without having to exchange the underlying datasets. FAMHE balances data protection, the accuracy or research results, and practical computational time. According to the team, FAMHE differs from other approaches in that it works at scale and has been mathematically proven to be secure. This article continues to discuss the importance of patient data privacy and the proposed FAMHE system, based on Multiparty Homomorphic Encryption (MHE), that enables privacy-preserving analyses of distributed datasets by producing accurate results without revealing intermediate data. 

Science Daily reports "A Cryptography Game-Changer for Biomedical Research at Scale"