JMIR Publications

ABSTRACT

Background:

Patient data is considered particularly sensitive personal data. Privacy regulation strictly governs their use and restricts their exchange. However, medical research can benefit from multicentric studies where patient data from different institutions is pooled and evaluated together. Here, the goals of data utilization and data protection are in conflict. Secure Multiparty Computation (SMPC) solves this conflict as it allows to directly compute on distributed proprietary data—held by different data owners—in a secure way without exchanging private data.

Objective:

The objective of this work is to provide a proof-of-principle of secure and privacy preserving multicentric computation by SMPC in the context of healthcare. To our knowledge, this is the first ever calculation with SMPC and real patient data over the free internet.

Methods:

The domain of survival analysis is particularly relevant in clinical research. For the Kaplan–Meier estimator we provide a secure version of the log-rank test. The complexity of the algorithm is explored both for synthetic data and for real patient data in a proof-of-principle over the Internet between Munich and Berlin.

Results:

We obtain a functional realization of an SMPC-based log-rank evaluation. This implementation is assessed with respect to performance and scaling behavior. We show that network latency strongly influences execution time of our solution. Furthermore, we identify a lower bound for the transmission rate which has to be fulfilled for unimpeded communication. In contrast, local performance of the participating parties have comparatively low influence on execution speed.

Conclusions:

We show that SMPC is applicable in the medical domain. A secure version of commonly used evaluation methods for clinical studies is possible with current implementations of SMPC. Furthermore, we derive that its application is practically feasible in terms of execution time.