Federated Learning of Electronic Health Records Improves Mortality Prediction in Patients Hospitalized with COVID-19

Abstract

Machine learning (ML) models require large datasets which may be siloed across different healthcare institutions. Using federated learning, a ML technique that avoids locally aggregating raw clinical data across multiple institutions, we predict mortality within seven days in hospitalized COVID-19 patients. Patient data was collected from Electronic Health Records (EHRs) from five hospitals within the Mount Sinai Health System (MSHS). Logistic Regression with L1 regularization (LASSO) and Multilayer Perceptron (MLP) models were trained using local data at each site, a pooled model with combined data from all five sites, and a federated model that only shared parameters with a central aggregator. Both the federated LASSO and federated MLP models performed better than their local model counterparts at four hospitals. The federated MLP model also outperformed the federated LASSO model at all hospitals. Federated learning shows promise in COVID-19 EHR data to develop robust predictive models without compromising patient privacy.

End-to-end privacy preserving deep learning on multi-institutional medical imaging

“…It has been proposed as a method to facilitate multi-national collaboration while obviating data transfer. In the setting of the COVID-19 pandemic 10,11 FL was used to allow the retention of data sovereignty and the enforcement of local governance policies over data repositories. In medical imaging, recent studies 5,12 demonstrated that federated training of deep learning models on brain tumour segmentation or breast density classification performs on-par with local training and that it fosters the inclusion of data from more diverse sources, leading to improved generalization.…”

On the Role of Artificial Intelligence in Medical Imaging of COVID-19

“…Federated learning (FL) is an emerging realm of ML concerned with distributed, decentralized training that stores privacy-sensitive data only locally (for details see 65,66,67 ). It allows multiple parties to collaboratively train the same model without data sharing and could thus become key to foster collaborations between clinical and AI communities and overcome privacy concerns.…”

“…Our meta-analysis included three preprints exploring FL using CT 68 or CXR 69 data. A recent FL study on EHR from 5 hospitals was found to improve COVID-19 mortality prediction 70 . These efforts will hopefully increase reproducibility and make comparative studies more feasible, which will help the research community focus on the highest performing methods.…”

Federated Learning used for predicting outcomes in SARS-COV-2 patients

“…The authors chose FL due to its ability to rapidly launch centrally orchestrated experiments with improved traceability of data and assessment of algorithmic changes and impact 29 . FL has shown promise in recent medical imaging applications [30][31][32][33] , including COVID-19 analysis [34][35][36][37] , albeit with limited scale. Governance of data for FL is maintained locally, alleviating privacy concerns, with only model 'weights' or 'gradients' transferred between the client-sites and the federated server 38,39 .…”