Frequency-mixed Single-source Domain Generalization for Medical Image Segmentation

TL;DR

This paper introduces FreeSDG, a novel method that enhances medical image segmentation models' ability to generalize to unseen domains using single-source data by frequency spectrum augmentation and self-supervision.

Contribution

The paper proposes FreeSDG, a new approach that uses mixed frequency spectrum and self-supervision to improve domain generalization from a single source in medical image segmentation.

Findings

Outperforms state-of-the-art methods on five datasets.

Significantly improves model generalizability to unseen domains.

Effective with limited annotated data.

Abstract

The annotation scarcity of medical image segmentation poses challenges in collecting sufficient training data for deep learning models. Specifically, models trained on limited data may not generalize well to other unseen data domains, resulting in a domain shift issue. Consequently, domain generalization (DG) is developed to boost the performance of segmentation models on unseen domains. However, the DG setup requires multiple source domains, which impedes the efficient deployment of segmentation algorithms in clinical scenarios. To address this challenge and improve the segmentation model's generalizability, we propose a novel approach called the Frequency-mixed Single-source Domain Generalization method (FreeSDG). By analyzing the frequency's effect on domain discrepancy, FreeSDG leverages a mixed frequency spectrum to augment the single-source domain. Additionally, self-supervision is constructed in the domain augmentation to learn robust context-aware representations for the segmentation task. Experimental results on five datasets of three modalities demonstrate the effectiveness of the proposed algorithm. FreeSDG outperforms state-of-the-art methods and significantly improves the segmentation model's generalizability. Therefore, FreeSDG provides a promising solution for enhancing the generalization of medical image segmentation models, especially when annotated data is scarce. The code is available at https://github.com/liamheng/Non-IID_Medical_Image_Segmentation.