The performance of deep learning-based methods strongly relies on the number of datasets used for training. Many efforts have been made to increase the data in the medical image analysis field. However, unlike photography images, it is hard to generate centralized databases to collect medical images because of numerous technical, legal, and privacy issues. In this work, we study the use of federated learning between two institutions in a real-world setting to collaboratively train a model without sharing the raw data across national boundaries. We quantitatively compare the segmentation models obtained with federated learning and local training alone. Our experimental results show that federated learning models have higher generalizability than standalone training.
@article{arxiv.2009.13148,
title = {Automated Pancreas Segmentation Using Multi-institutional Collaborative Deep Learning},
author = {Pochuan Wang and Chen Shen and Holger R. Roth and Dong Yang and Daguang Xu and Masahiro Oda and Kazunari Misawa and Po-Ting Chen and Kao-Lang Liu and Wei-Chih Liao and Weichung Wang and Kensaku Mori},
journal= {arXiv preprint arXiv:2009.13148},
year = {2020}
}