Distributed deep learning networks among institutions for medical imaging

Chang, Ken and Balachandar, Niranjan and Lam, Carson and Yi, Darvin and Brown, James and Beers, Andrew and Rosen, Bruce and Rubin, Daniel L and Kalpathy-Cramer, Jayashree (2018) Distributed deep learning networks among institutions for medical imaging. Journal of the American Medical Informatics Association, 25 (8). pp. 945-954. ISSN 1067-5027

Full content URL: https://doi.org/10.1093/jamia/ocy017

Documents
Distributed deep learning networks among institutions for medical imaging
Publishers PDF
[img]
[Download]
[img] PDF
JB Pub PDF.pdf - Whole Document
Available under License Creative Commons Attribution-NonCommercial 4.0 International.

642kB
Item Type:Article
Item Status:Live Archive

Abstract

Objective
Deep learning has become a promising approach for automated support for clinical diagnosis. When medical data samples are limited, collaboration among multiple institutions is necessary to achieve high algorithm performance. However, sharing patient data often has limitations due to technical, legal, or ethical concerns. In this study, we propose methods of distributing deep learning models as an attractive alternative to sharing patient data.

Methods
We simulate the distribution of deep learning models across 4 institutions using various training heuristics and compare the results with a deep learning model trained on centrally hosted patient data. The training heuristics investigated include ensembling single institution models, single weight transfer, and cyclical weight transfer. We evaluated these approaches for image classification in 3 independent image collections (retinal fundus photos, mammography, and ImageNet).

Results
We find that cyclical weight transfer resulted in a performance that was comparable to that of centrally hosted patient data. We also found that there is an improvement in the performance of cyclical weight transfer heuristic with a high frequency of weight transfer.

Conclusions
We show that distributing deep learning models is an effective alternative to sharing patient data. This finding has implications for any collaborative deep learning study.

Keywords:deep learning, neural networks, distributed learning, medical imaging
Subjects:G Mathematical and Computer Sciences > G730 Neural Computing
G Mathematical and Computer Sciences > G760 Machine Learning
Divisions:College of Science > School of Computer Science
ID Code:35639
Deposited On:15 Apr 2019 08:16

Repository Staff Only: item control page