# UCAD: Uncertainty-guided Contour-aware Displacement for semi-supervised medical image segmentation

Source: [UCAD: Uncertainty-guided Contour-aware Displacement for semi-supervised medical image segmentation](<https://www.sota2.com/research/paper/69784664a2b28e761d87c357>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-28T05:00:08.385Z

Authors: Chengbo Ding, Fenghe Tang, Shaohua Kevin Zhou

Published: 2026-01-24T08:21:14.000Z

## Abstract

Existing displacement strategies in semi-supervised segmentation only operate on rectangular regions, ignoring anatomical structures and resulting in boundary distortions and semantic inconsistency. To address these issues, we propose UCAD, an Uncertainty-Guided Contour-Aware Displacement framework for semi-supervised medical image segmentation that preserves contour-aware semantics while enhancing consistency learning. Our UCAD leverages superpixels to generate anatomically coherent regions aligned with anatomy boundaries, and an uncertainty-guided selection mechanism to selectively displace challenging regions for better consistency learning. We further propose a dynamic uncertainty-weighted consistency loss, which adaptively stabilizes training and effectively regularizes the model on unlabeled regions. Extensive experiments demonstrate that UCAD consistently outperforms state-of-the-art semi-supervised segmentation methods, achieving superior segmentation accuracy under limited annotation. The code is available at:https://github.com/dcb937/UCAD.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Semi-supervised Medical Image Segmentation on ACDC (5% labeled)](<https://www.sota2.com/research/sota/semi-supervised-medical-image-segmentation-on-acdc-5-percent-labeled.md>): Semi-supervised medical image segmentation; ACDC (5% labeled); DSC: 88.63.

- [Semi-supervised Medical Image Segmentation on ACDC 10% labeled](<https://www.sota2.com/research/sota/semi-supervised-medical-image-segmentation-on-acdc-10-percent-labeled.md>): Semi-supervised medical image segmentation; ACDC 10% labeled; DSC: 89.93.

- [Semi-supervised medical image segmentation on Synapse (5% label)](<https://www.sota2.com/research/sota/semi-supervised-medical-image-segmentation-on-synapse-5-percent-label.md>): Semi-supervised medical image segmentation; Synapse (5% label); DSC: 56.1.

- [Semi-supervised medical image segmentation on Synapse 10% label](<https://www.sota2.com/research/sota/semi-supervised-medical-image-segmentation-on-synapse-10-percent-label.md>): Semi-supervised medical image segmentation; Synapse 10% label; DSC: 0.6673.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2601.17366>)

[PDF](<https://arxiv.org/pdf/2601.17366.pdf>)
