Federated Analysis of Neuroimaging Data Using Coinstac

Federated analysis can help perform large-scale analyses using neuroimaging datasets across various research groups overcoming the limitations of institutional data-sharing poli-cies, privacy or regulatory concerns as it requires no data sharing. In this work, we showcase our federated platform, COINSTAC, a fully secure, privacy-preserving neuroinfor-matics platform that allows large-scale analysis of federated neuroimaging data (no sharing) and offers several feder-ated algorithms with results on par with centralized analysis (shared/pooled). All federated computations (applications) currently existing in the COINSTAC are trained and tested to ensure that the results of these federated methods perform similarly to their corresponding centralized methods.