Bridging spatial awareness and global context in medical image segmentation

2025-12-06

AI Summary

Medical image segmentation suffers from an imbalance between local detail preservation and global context modeling, leading to ambiguous boundaries and structural omissions. To address this, we propose U-CycleMLP—a U-shaped architecture integrating dense atrous convolutions, multi-level down/up-sampling, and skip connections. We introduce two key innovations: (1) a positional attention weight excitation module for spatially adaptive feature recalibration, and (2) channel-wise CycleMLP blocks enabling efficient, linear-complexity (O(N)) multi-scale feature fusion and boundary refinement. Evaluated on three major public benchmarks, U-CycleMLP achieves state-of-the-art performance, particularly excelling in fine-grained anatomical structure segmentation and cross-modality robustness (CT/MRI). The framework establishes a new paradigm for efficient and precise medical image segmentation.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• User Modeling, Personalization and Recommendation: ML for personalized search and recommendations
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs

Abstract

Medical image segmentation is a fundamental task in computer-aided diagnosis, requiring models that balance segmentation accuracy and computational efficiency. However, existing segmentation models often struggle to effectively capture local and global contextual information, leading to boundary pixel loss and segmentation errors. In this paper, we propose U-CycleMLP, a novel U-shaped encoder-decoder network designed to enhance segmentation performance while maintaining a lightweight architecture. The encoder learns multiscale contextual features using position attention weight excitation blocks, dense atrous blocks, and downsampling operations, effectively capturing both local and global contextual information. The decoder reconstructs high-resolution segmentation masks through upsampling operations, dense atrous blocks, and feature fusion mechanisms, ensuring precise boundary delineation. To further refine segmentation predictions, channel CycleMLP blocks are incorporated into the decoder along the skip connections, enhancing feature integration while maintaining linear computational complexity relative to input size. Experimental results, both quantitative and qualitative, across three benchmark datasets demonstrate the competitive performance of U-CycleMLP in comparison with state-of-the-art methods, achieving better segmentation accuracy across all datasets, capturing fine-grained anatomical structures, and demonstrating robustness across different medical imaging modalities. Ablation studies further highlight the importance of the model's core architectural components in enhancing segmentation accuracy.

Problem

• Enhance medical image segmentation accuracy and efficiency
• Capture both local and global contextual information effectively
• Maintain lightweight architecture while improving boundary delineation

Innovation

• U-shaped encoder-decoder network with lightweight design
• Position attention and dense atrous blocks for multiscale context
• Channel CycleMLP blocks for efficient feature integration