1. Federated learning improves site performance in multicenter deep learning without data sharing
- Author
-
Leonard S. Marks, William Speier, Baris Turkbey, Bradford J. Wood, Steven S. Raman, Thomas Sanford, Rushikesh Kulkarni, Alex G. Raman, Jesse Tetreault, Alan Priester, Mona Flores, Holger R. Roth, Daguang Xu, Ziyue Xu, Peter L. Choyke, Corey W. Arnold, Karthik V. Sarma, Stephanie Harmon, and Dieter R. Enzmann
- Subjects
AcademicSubjects/SCI01060 ,Pooling ,Library science ,Health Informatics ,02 engineering and technology ,privacy ,Medical and Health Sciences ,Federated learning ,030218 nuclear medicine & medical imaging ,03 medical and health sciences ,Deep Learning ,Engineering ,0302 clinical medicine ,Information and Computing Sciences ,0202 electrical engineering, electronic engineering, information engineering ,Humans ,Generalizability theory ,Overall performance ,generalizability ,AcademicSubjects/MED00580 ,prostate ,federated learning ,Information Dissemination ,business.industry ,Deep learning ,deep learning ,Learning models ,Test (assessment) ,Data sharing ,Privacy ,020201 artificial intelligence & image processing ,Artificial intelligence ,AcademicSubjects/SCI01530 ,Brief Communications ,business ,Psychology ,Medical Informatics - Abstract
Objective To demonstrate enabling multi-institutional training without centralizing or sharing the underlying physical data via federated learning (FL). Materials and Methods Deep learning models were trained at each participating institution using local clinical data, and an additional model was trained using FL across all of the institutions. Results We found that the FL model exhibited superior performance and generalizability to the models trained at single institutions, with an overall performance level that was significantly better than that of any of the institutional models alone when evaluated on held-out test sets from each institution and an outside challenge dataset. Discussion The power of FL was successfully demonstrated across 3 academic institutions while avoiding the privacy risk associated with the transfer and pooling of patient data. Conclusion Federated learning is an effective methodology that merits further study to enable accelerated development of models across institutions, enabling greater generalizability in clinical use.
- Published
- 2021