Unleashing the Power of Intermediate Domains for Mixed Domain Semi-Supervised Medical Image Segmentation

TL;DR

This paper introduces UST-RUN, a novel framework that leverages intermediate domain information to improve semi-supervised medical image segmentation across multiple domains, addressing both limited annotations and domain shift.

Contribution

The paper proposes a new approach combining intermediate domain construction, symmetric guidance, and style-transfer techniques to enhance knowledge transfer in mixed domain semi-supervised segmentation.

Findings

Achieves 12.94% improvement in Dice score on Prostate dataset.

Outperforms existing methods on four public datasets.

Effectively handles domain shift and limited annotations.

Abstract

Both limited annotation and domain shift are prevalent challenges in medical image segmentation. Traditional semi-supervised segmentation and unsupervised domain adaptation methods address one of these issues separately. However, the coexistence of limited annotation and domain shift is quite common, which motivates us to introduce a novel and challenging scenario: Mixed Domain Semi-supervised medical image Segmentation (MiDSS), where limited labeled data from a single domain and a large amount of unlabeled data from multiple domains. To tackle this issue, we propose the UST-RUN framework, which fully leverages intermediate domain information to facilitate knowledge transfer. We employ Unified Copy-paste (UCP) to construct intermediate domains, and propose a Symmetric GuiDance training strategy (SymGD) to supervise unlabeled data by merging pseudo-labels from intermediate samples. Subsequently, we introduce a Training Process aware Random Amplitude MixUp (TP-RAM) to progressively incorporate style-transition components into intermediate samples. To generate more diverse intermediate samples, we further select reliable samples with high-quality pseudo-labels, which are then mixed with other unlabeled data. Additionally, we generate sophisticated intermediate samples with high-quality pseudo-labels for unreliable samples, ensuring effective knowledge transfer for them. Extensive experiments on four public datasets demonstrate the superiority of UST-RUN. Notably, UST-RUN achieves a 12.94% improvement in Dice score on the Prostate dataset. Our code is available at https://github.com/MQinghe/UST-RUN