Multiscale Switch for Semi-Supervised and Contrastive Learning in Medical Ultrasound Image Segmentation

TL;DR

This paper introduces Switch, a semi-supervised ultrasound image segmentation framework that uses multiscale patch mixing and Fourier domain contrastive learning to improve feature robustness and segmentation accuracy with limited labeled data.

Contribution

The paper proposes a novel SSL framework with hierarchical patch mixing and Fourier domain contrastive learning, enhancing unlabeled data utilization and feature representation in medical ultrasound segmentation.

Findings

Achieves over 80% Dice on lymph node dataset at 5% labels

Outperforms state-of-the-art semi-supervised methods across six datasets

Maintains parameter efficiency with 1.8M parameters

Abstract

Medical ultrasound image segmentation faces significant challenges due to limited labeled data and characteristic imaging artifacts including speckle noise and low-contrast boundaries. While semi-supervised learning (SSL) approaches have emerged to address data scarcity, existing methods suffer from suboptimal unlabeled data utilization and lack robust feature representation mechanisms. In this paper, we propose Switch, a novel SSL framework with two key innovations: (1) Multiscale Switch (MSS) strategy that employs hierarchical patch mixing to achieve uniform spatial coverage; (2) Frequency Domain Switch (FDS) with contrastive learning that performs amplitude switching in Fourier space for robust feature representations. Our framework integrates these components within a teacher-student architecture to effectively leverage both labeled and unlabeled data. Comprehensive evaluation across six diverse ultrasound datasets (lymph nodes, breast lesions, thyroid nodules, and prostate) demonstrates consistent superiority over state-of-the-art methods. At 5\% labeling ratio, Switch achieves remarkable improvements: 80.04\% Dice on LN-INT, 85.52\% Dice on DDTI, and 83.48\% Dice on Prostate datasets, with our semi-supervised approach even exceeding fully supervised baselines. The method maintains parameter efficiency (1.8M parameters) while delivering superior performance, validating its effectiveness for resource-constrained medical imaging applications. The source code is publicly available at https://github.com/jinggqu/Switch