MIS-HCC: Hierarchical Channel Clustering for Efficient Medical Image Segmentation

2026-07-19

AI Summary

This work addresses the challenge of deploying medical image segmentation models on resource-constrained devices, where high computational cost and parameter redundancy hinder practical application. Existing compression methods often overlook intra-layer channel structural similarities, leading to suboptimal preservation of critical features. To overcome this limitation, the authors propose a hierarchical clustering–based channel compression approach that, for the first time, incorporates the Wasserstein distance into medical image segmentation model compression. By measuring distributional similarity among channels in a pretrained network, a similarity matrix is constructed to guide hierarchical clustering and subsequent parameter fusion of clustered channels. Evaluated on three medical image datasets, the method consistently outperforms state-of-the-art techniques, achieving superior compression efficiency while maintaining high segmentation accuracy, thereby offering an improved trade-off between model size and performance suitable for edge-based medical deployment scenarios.

Abstract

Medical image segmentation models require both high accuracy and lightweight design to accommodate real-world medical applications. The deployment of these models on resource-limited medical platforms remains a significant challenge due to their high computational and parameter requirements. Existing pruning methods for model compression mostly overlook the intrinsic connections and similarity between the internal structures of complex deep neural networks. As a result, compressed models may not effectively retain the basic features of the pretrained network. To solve this problem, we propose a hierarchical clustering compression method for medical image segmentation models (MIS-HCC). This approach employs hierarchical clustering to partition channels and fuse their parameters efficiently. Specifically, it leverages the Wasserstein distance to represent similarity of channels within layers of pre-trained network, forming a similarity matrix that guides the clustering process. Channels within each cluster are then fused to produce a compressed network. Experimental results on three medical image datasets application demonstrate that MIS-HCC outperforms the state-of-the-art methods in both accuracy and compression efficiency, offering an effective solution for deploying medical image segmentation models on resource-limited medical platforms.

Problem

• medical image segmentation
• model compression
• channel clustering
• resource-limited deployment
• deep neural networks

Innovation

• hierarchical clustering
• channel fusion
• Wasserstein distance
• model compression
• medical image segmentation