Distributed deep learning networks among institutions for medical imaging

By Ken Chang, Niranjan Balachandar, Carson K. Lam, Darvin Yi, James M. Brown, Andrew L Beers, Bruce R. Rosen, Daniel L. Rubin, Jayashree Kalpathy-Cramer. 2018

315 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Health Informatics, Radiology, Nuclear Medicine and Imaging, Medical Imaging, Federated & Distributed Learning

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Health Informatics

In Radiology, Nuclear Medicine and Imaging

In Medical Imaging

In Federated & Distributed Learning

Citations in the corpus

Advances and Open Problems in Federated Learning (2019, background): “ modeling occurring in institutional silos: for example, using electronic health records or clinical images from individual medical institutions instead of pooling data and models across institutions [83, 93]. In contexts where membership in institutional datasets is correlated with individuals’ spe”
An overview of deep learning in medical imaging focusing on MRI (2018, background): “There are a few examples of these ideas entering the medical machine learning community, as in [336] where the distribution of deep learning models among several medical institutions was investigated, but then without considering the above privacy issues.”
Artificial intelligence in Cancer Imaging: Clinical Challenges and Applications (2019, background): “207 For example, the National Institutes of Health recently shared chest x-ray and CT repositories to help AI scientists.”
Preparing Medical Imaging Data for Machine Learning (2020, background): “Large sample sizes are especially required in populations with substantial heterogeneity or when differences between imaging phenotypes are subtle (35).”
Privacy-Preserving Distributed Deep Learning for Clinical Data (2018, methodology): “In particular, we incorporate differential privacy with cyclical weight transfer [Chang et al., 2018], by training deep neural networks under differential privacy in a distributed fashion.”
Decentralized Federated Learning: A Survey and Perspective (2023, methodology): “The line variant is frequently used as a baseline for comparison due to its ease of implementation, intuitiveness, and efﬁciency [163], [164], [165], [166].”
Federated learning in medicine: facilitating multi-institutional collaborations without sharing patient data (2020, methodology): “We stress that CIIL “best local” requires more communication between institutions than was originally designed for 14 .”
Federated learning enables big data for rare cancer boundary detection (2022, background): “The exact 49 site IDs are: 1, 2, 3, 4, 5, 6, 7, 9, 10, 12, 13, 14, 15, 16, 17, 18, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 44, 45, 46, 48, 49, 50, 52, 53, 54, 56, 59, 60.”
Artificial intelligence and machine learning for medical imaging: A technology review. (2021, unknown): “In addition, federated learning approaches [235 – 237] can be used to train AI models across institutions while ensuring data privacy protection, sharing the clinical knowledge and getting the advantages of collaborative AI solutions.”
Federated learning and differential privacy for medical image analysis (2021, background): “Previous studies have mostly experimented with a fixed number of clients having similar distributions of data 1,13,36 .”
Model Aggregation via Good-Enough Model Spaces (2018, background): “, a network of different hospitals) [3].”
ExpertMatcher: Automating ML Model Selection for Users in Resource Constrained Countries (2019, methodology): “Thanks to modern cloud infrastructure and distributed learning methods (Vepakomma et al. , 2018b, 2019, 2018a; Chang et al. , 2018), “expert model hubs” can make inference available to any remote client connected to the internet.”