Bidirectional Channel-selective Semantic Interaction for Semi-Supervised Medical Segmentation

2026-01-09

arXiv.org

AI Summary

This work proposes the BCSI framework to address key challenges in semi-supervised medical image segmentation, including scarce annotated data, error accumulation, and insufficient interaction between labeled and unlabeled samples. BCSI enhances data consistency through semantic-spatial perturbation (SSP) and introduces two novel mechanisms: channel-selective routing (CR) and bidirectional channel interaction (BCI), which jointly optimize semantic information exchange between labeled and unlabeled data. These components effectively suppress noise interference and strengthen the representation of critical features. Extensive experiments on multiple 3D medical image benchmark datasets demonstrate that BCSI significantly outperforms existing methods, achieving superior segmentation accuracy and strong generalization capability.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Semi-Supervised Learning
• Reasoning under Uncertainty: Uncertainty Representations

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

Semi-supervised medical image segmentation is an effective method for addressing scenarios with limited labeled data. Existing methods mainly rely on frameworks such as mean teacher and dual-stream consistency learning. These approaches often face issues like error accumulation and model structural complexity, while also neglecting the interaction between labeled and unlabeled data streams. To overcome these challenges, we propose a Bidirectional Channel-selective Semantic Interaction~(BCSI) framework for semi-supervised medical image segmentation. First, we propose a Semantic-Spatial Perturbation~(SSP) mechanism, which disturbs the data using two strong augmentation operations and leverages unsupervised learning with pseudo-labels from weak augmentations. Additionally, we employ consistency on the predictions from the two strong augmentations to further improve model stability and robustness. Second, to reduce noise during the interaction between labeled and unlabeled data, we propose a Channel-selective Router~(CR) component, which dynamically selects the most relevant channels for information exchange. This mechanism ensures that only highly relevant features are activated, minimizing unnecessary interference. Finally, the Bidirectional Channel-wise Interaction~(BCI) strategy is employed to supplement additional semantic information and enhance the representation of important channels. Experimental results on multiple benchmarking 3D medical datasets demonstrate that the proposed method outperforms existing semi-supervised approaches.

Problem

• semi-supervised medical segmentation
• error accumulation
• model complexity
• labeled-unlabeled data interaction
• channel-wise interaction

Innovation

• Bidirectional Channel-selective Semantic Interaction
• Semantic-Spatial Perturbation
• Channel-selective Router
• Semi-supervised Medical Segmentation
• Consistency Learning