Segmentation of Multiple Sclerosis (MS) lesions is a challenging problem. Several deep-learning-based methods have been proposed in recent years. However, most methods tend to be static, that is, a single model trained on a large, specialized dataset, which does not generalize well. Instead, the model should learn across datasets arriving sequentially from different hospitals by building upon the characteristics of lesions in a continual manner. In this regard, we explore experience replay, a well-known continual learning method, in the context of MS lesion segmentation across multi-contrast data from 8 different hospitals. Our experiments show that replay is able to achieve positive backward transfer and reduce catastrophic forgetting compared to sequential fine-tuning. Furthermore, replay outperforms the multi-domain training, thereby emerging as a promising solution for the segmentation of MS lesions. The code is available at this link: https://github.com/naga-karthik/continual-learning-ms
@article{arxiv.2210.15091,
title = {Segmentation of Multiple Sclerosis Lesions across Hospitals: Learn Continually or Train from Scratch?},
author = {Enamundram Naga Karthik and Anne Kerbrat and Pierre Labauge and Tobias Granberg and Jason Talbott and Daniel S. Reich and Massimo Filippi and Rohit Bakshi and Virginie Callot and Sarath Chandar and Julien Cohen-Adad},
journal= {arXiv preprint arXiv:2210.15091},
year = {2022}
}
Comments
Accepted at the Medical Imaging Meets NeurIPS (MedNeurIPS) Workshop 2022