Secure Federated Learning for Clinical Informatics with Applications to the COVID-19 Pandemic

Enabling health care providers to respond faster and with greater precision to pandemics requires both advanced machine learning and quickly accessible clinical data. Yet, the necessary medical data is often inaccessible across hospitals due to privacy and intellectual property concerns. This proposal leverages distributed machine learning and modern cryptography to introduce a computational protocol and software tools for securely training machine learning models with data spread over several medical establishments, while preserving privacy and IP rights. Our scientific contributions include innovative techniques that trade-off computation and communication to improve the predictive performance of federated learning in clinical settings and novel cryptographic techniques that trade off computation and robustness to enhance security. To complement our technical aims, we will develop open source software. We will evaluate our approach for COVID diagnosis using data available on the C3.ai Data Lake combined with clinical data from OSF HealthCare, to illustrate how private data can significantly improve prediction quality compared to public data alone. We also propose to serve as a hub for other c3 AI projects to enable the secure use of privately-held clinical datasets, which will improve results by other teams. Our broader vision and objective are to provide Secure Federated Learning as a Service (FLaaS) freely available to any hospital during a declared crisis. We envision that a robust, secure federated learning system will enable fast responses to minimize the impact of disease in the earliest stages.