medRxiv · 2020 · 50 citations · 18 references
ComputingBiomedical InformaticsHealth InformaticsElectronic Health RecordsMachine LearningPredictive AnalyticsElectronic Health Record AnalyticsElectronic Health Record PrivacyFederated LearningHealthcare AnalyticsMachine Learning For Big DataPredictive Health ModelingElectronic Health Record Prediction ModelsHospital Electronic Health RecordsL1 RegularizationLassoLogistic RegressionMachine Learning For HealthcareMachine Learning Prediction ModelsRegularization
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.
18
Stephen A. Lauer, Kyra H. Grantz, Qifang Bi et al. · Annals of Internal Medicine · 2020 · 6.6K citations · Full text
Veterinary Medicine, Infectious Disease Control, Infectious Diseases +17
CT Imaging Features of 2019 Novel Coronavirus (2019-nCoV)
Michael Chung, Adam Bernheim, Xueyan Mei et al. · Radiology · 2020 · 2.9K citations · Full text
Federated Learning for Healthcare Informatics
Jie Xu, Benjamin S. Glicksberg, Chang Su et al. · Journal of Healthcare Informatics Research · 2020 · 1.4K citations · Full text