CAD: Confidence-Aware Adaptive Displacement for Semi-Supervised Medical Image Segmentation

TL;DR

This paper introduces CAD, a confidence-aware framework for semi-supervised medical image segmentation that adaptively refines predictions by replacing uncertain regions with high-confidence patches, improving accuracy.

Contribution

CAD is a novel adaptive method that selectively replaces low-confidence regions with high-confidence patches, dynamically adjusting thresholds to enhance segmentation quality.

Findings

CAD achieves state-of-the-art accuracy on public datasets.

The adaptive displacement strategy improves segmentation consistency.

The method effectively handles uncertain regions in semi-supervised learning.

Abstract

Semi-supervised medical image segmentation aims to leverage minimal expert annotations, yet remains confronted by challenges in maintaining high-quality consistency learning. Excessive perturbations can degrade alignment and hinder precise decision boundaries, especially in regions with uncertain predictions. In this paper, we introduce Confidence-Aware Adaptive Displacement (CAD), a framework that selectively identifies and replaces the largest low-confidence regions with high-confidence patches. By dynamically adjusting both the maximum allowable replacement size and the confidence threshold throughout training, CAD progressively refines the segmentation quality without overwhelming the learning process. Experimental results on public medical datasets demonstrate that CAD effectively enhances segmentation quality, establishing new state-of-the-art accuracy in this field. The source code will be released after the paper is published.