Towards a Collective Medical Imaging AI: Enabling Continual Learning from Peers

Guangyao Zheng, Vladimir Braverman, Jeffrey Leal, Steven Rowe, Doris Leung, Michael A. Jacobs, Vishwa Sanjay Parekh
Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning, PMLR 250:1862-1877, 2024.

Abstract

Federated learning is an exciting area within machine learning that allows cross-silo training of large-scale machine learning models on disparate or similar tasks in a privacy-preserving manner. However, conventional federated learning frameworks require a synchronous training schedule and are incapable of lifelong learning. To that end, we propose an asynchronous decentralized federated lifelong learning (ADFLL) method that allows agents in the system to asynchronously and continually learn from their own previous experiences and others\’, thus overcoming the potential drawbacks of conventional federated learning. We evaluate the ADFLL framework in two experimental setups for deep reinforcement learning (DRL) based landmark localization across different imaging modalities, orientations, and sequences. The ADFLL was compared to central aggregation and conventional lifelong learning for upper-bound comparison and with a conventional DRL model for lower-bound comparison. Across all the experiments, ADFLL demonstrated excellent capability to collaboratively learn all tasks across all the agents compared to the baseline models in in-distribution and out-of-distribution test sets. In conclusion, we provide a flexible, efficient, and robust federated lifelong learning framework that can be deployed in real-world applications.

Cite this Paper

BibTeX
@InProceedings{pmlr-v250-zheng24a, title = {Towards a Collective Medical Imaging AI: Enabling Continual Learning from Peers}, author = {Zheng, Guangyao and Braverman, Vladimir and Leal, Jeffrey and Rowe, Steven and Leung, Doris and Jacobs, Michael A. and Parekh, Vishwa Sanjay}, booktitle = {Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning}, pages = {1862--1877}, year = {2024}, editor = {Burgos, Ninon and Petitjean, Caroline and Vakalopoulou, Maria and Christodoulidis, Stergios and Coupe, Pierrick and Delingette, Hervé and Lartizien, Carole and Mateus, Diana}, volume = {250}, series = {Proceedings of Machine Learning Research}, month = {03--05 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v250/main/assets/zheng24a/zheng24a.pdf}, url = {https://proceedings.mlr.press/v250/zheng24a.html}, abstract = {Federated learning is an exciting area within machine learning that allows cross-silo training of large-scale machine learning models on disparate or similar tasks in a privacy-preserving manner. However, conventional federated learning frameworks require a synchronous training schedule and are incapable of lifelong learning. To that end, we propose an asynchronous decentralized federated lifelong learning (ADFLL) method that allows agents in the system to asynchronously and continually learn from their own previous experiences and others\’, thus overcoming the potential drawbacks of conventional federated learning. We evaluate the ADFLL framework in two experimental setups for deep reinforcement learning (DRL) based landmark localization across different imaging modalities, orientations, and sequences. The ADFLL was compared to central aggregation and conventional lifelong learning for upper-bound comparison and with a conventional DRL model for lower-bound comparison. Across all the experiments, ADFLL demonstrated excellent capability to collaboratively learn all tasks across all the agents compared to the baseline models in in-distribution and out-of-distribution test sets. In conclusion, we provide a flexible, efficient, and robust federated lifelong learning framework that can be deployed in real-world applications.} }
Endnote
%0 Conference Paper %T Towards a Collective Medical Imaging AI: Enabling Continual Learning from Peers %A Guangyao Zheng %A Vladimir Braverman %A Jeffrey Leal %A Steven Rowe %A Doris Leung %A Michael A. Jacobs %A Vishwa Sanjay Parekh %B Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2024 %E Ninon Burgos %E Caroline Petitjean %E Maria Vakalopoulou %E Stergios Christodoulidis %E Pierrick Coupe %E Hervé Delingette %E Carole Lartizien %E Diana Mateus %F pmlr-v250-zheng24a %I PMLR %P 1862--1877 %U https://proceedings.mlr.press/v250/zheng24a.html %V 250 %X Federated learning is an exciting area within machine learning that allows cross-silo training of large-scale machine learning models on disparate or similar tasks in a privacy-preserving manner. However, conventional federated learning frameworks require a synchronous training schedule and are incapable of lifelong learning. To that end, we propose an asynchronous decentralized federated lifelong learning (ADFLL) method that allows agents in the system to asynchronously and continually learn from their own previous experiences and others\’, thus overcoming the potential drawbacks of conventional federated learning. We evaluate the ADFLL framework in two experimental setups for deep reinforcement learning (DRL) based landmark localization across different imaging modalities, orientations, and sequences. The ADFLL was compared to central aggregation and conventional lifelong learning for upper-bound comparison and with a conventional DRL model for lower-bound comparison. Across all the experiments, ADFLL demonstrated excellent capability to collaboratively learn all tasks across all the agents compared to the baseline models in in-distribution and out-of-distribution test sets. In conclusion, we provide a flexible, efficient, and robust federated lifelong learning framework that can be deployed in real-world applications.
APA
Zheng, G., Braverman, V., Leal, J., Rowe, S., Leung, D., Jacobs, M.A. & Parekh, V.S.. (2024). Towards a Collective Medical Imaging AI: Enabling Continual Learning from Peers. Proceedings of The 7nd International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 250:1862-1877 Available from https://proceedings.mlr.press/v250/zheng24a.html.

Related Material