SL: Stable Learning in Source-Free Domain Adaption for Medical Image Segmentation

TL;DR

This paper introduces a Stable Learning strategy for source-free unsupervised domain adaptation in medical image segmentation, addressing overfitting issues and improving performance without access to source data.

Contribution

The paper proposes a novel Stable Learning approach with Weight Consolidation and Entropy Increase to prevent overfitting in source-free domain adaptation for medical imaging.

Findings

SL improves segmentation accuracy in source-free UDA scenarios.

The method effectively mitigates overfitting during training.

Experimental results outperform existing methods.

Abstract

Deep learning techniques for medical image analysis usually suffer from the domain shift between source and target data. Most existing works focus on unsupervised domain adaptation (UDA). However, in practical applications, privacy issues are much more severe. For example, the data of different hospitals have domain shifts due to equipment problems, and data of the two domains cannot be available simultaneously because of privacy. In this challenge defined as Source-Free UDA, the previous UDA medical methods are limited. Although a variety of medical source-free unsupervised domain adaption (MSFUDA) methods have been proposed, we found they fall into an over-fitting dilemma called "longer training, worse performance." Therefore, we propose the Stable Learning (SL) strategy to address the dilemma. SL is a scalable method and can be integrated with other research, which consists of Weight Consolidation and Entropy Increase. First, we apply Weight Consolidation to retain domain-invariant knowledge and then we design Entropy Increase to avoid over-learning. Comparative experiments prove the effectiveness of SL. We also have done extensive ablation experiments. Besides, We will release codes including a variety of MSFUDA methods.