TY  - JOUR
AU  - Vaid, Akhil
AU  - Jaladanki, Suraj K
AU  - Xu, Jie
AU  - Teng, Shelly
AU  - Kumar, Arvind
AU  - Lee, Samuel
AU  - Somani, Sulaiman
AU  - Paranjpe, Ishan
AU  - De Freitas, Jessica K
AU  - Wanyan, Tingyi
AU  - Johnson, Kipp W
AU  - Bicak, Mesude
AU  - Klang, Eyal
AU  - Kwon, Young Joon
AU  - Costa, Anthony
AU  - Zhao, Shan
AU  - Miotto, Riccardo
AU  - Charney, Alexander W
AU  - Böttinger, Erwin
AU  - Fayad, Zahi A
AU  - Nadkarni, Girish N
AU  - Wang, Fei
AU  - Glicksberg, Benjamin S
PY  - 2021
DA  - 2021/1/27
TI  - Federated Learning of Electronic Health Records to Improve Mortality Prediction in Hospitalized Patients With COVID-19: Machine Learning Approach
JO  - JMIR Med Inform
SP  - e24207
VL  - 9
IS  - 1
KW  - federated learning
KW  - COVID-19
KW  - machine learning
KW  - electronic health records
AB  - Background: Machine learning models require large datasets that may be siloed across different health care institutions. Machine learning studies that focus on COVID-19 have been limited to single-hospital data, which limits model generalizability. Objective: We aimed to use federated learning, a machine learning technique that avoids locally aggregating raw clinical data across multiple institutions, to predict mortality in hospitalized patients with COVID-19 within 7 days. Methods: Patient data were collected from the electronic health records of 5 hospitals within the Mount Sinai Health System. Logistic regression with L1 regularization/least absolute shrinkage and selection operator (LASSO) and multilayer perceptron (MLP) models were trained by using local data at each site. We developed a pooled model with combined data from all 5 sites, and a federated model that only shared parameters with a central aggregator. Results: The LASSOfederated model outperformed the LASSOlocal model at 3 hospitals, and the MLPfederated model performed better than the MLPlocal model at all 5 hospitals, as determined by the area under the receiver operating characteristic curve. The LASSOpooled model outperformed the LASSOfederated model at all hospitals, and the MLPfederated model outperformed the MLPpooled model at 2 hospitals. Conclusions: The federated learning of COVID-19 electronic health record data shows promise in developing robust predictive models without compromising patient privacy. 
SN  - 2291-9694
UR  - http://medinform.jmir.org/2021/1/e24207/
UR  - https://doi.org/10.2196/24207
UR  - http://www.ncbi.nlm.nih.gov/pubmed/33400679
DO  - 10.2196/24207
ID  - info:doi/10.2196/24207
ER  -