Focal Modulation and Bidirectional Feature Fusion Network for Medical Image Segmentation

2025-10-23

AI Summary

Medical image segmentation faces two key challenges: insufficient global context modeling and difficulty in representing multi-scale anatomical structures. Conventional CNNs, constrained by local receptive fields, struggle to accurately segment lesions with complex boundaries or highly variable sizes. To address these limitations, we propose a convolutional-Transformer hybrid architecture featuring: (1) focal modulation attention, which enhances long-range dependency modeling and selective focus on discriminative regions; and (2) a bidirectional cross-scale feature fusion module enabling efficient, symmetric information exchange between encoder and decoder pathways. Evaluated on eight benchmark medical image segmentation datasets, our method achieves state-of-the-art performance, consistently outperforming existing approaches in both Jaccard and Dice coefficients. Ablation studies confirm its superior boundary precision, scale robustness, and generalization capability across diverse anatomical domains and imaging modalities.

Technology Category

• Computer Vision: Segmentation
• Machine Learning: Multimodal Learning
• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Security and Privacy: Data transparency and provenance

Abstract

Medical image segmentation is essential for clinical applications such as disease diagnosis, treatment planning, and disease development monitoring because it provides precise morphological and spatial information on anatomical structures that directly influence treatment decisions. Convolutional neural networks significantly impact image segmentation; however, since convolution operations are local, capturing global contextual information and long-range dependencies is still challenging. Their capacity to precisely segment structures with complicated borders and a variety of sizes is impacted by this restriction. Since transformers use self-attention methods to capture global context and long-range dependencies efficiently, integrating transformer-based architecture with CNNs is a feasible approach to overcoming these challenges. To address these challenges, we propose the Focal Modulation and Bidirectional Feature Fusion Network for Medical Image Segmentation, referred to as FM-BFF-Net in the remainder of this paper. The network combines convolutional and transformer components, employs a focal modulation attention mechanism to refine context awareness, and introduces a bidirectional feature fusion module that enables efficient interaction between encoder and decoder representations across scales. Through this design, FM-BFF-Net enhances boundary precision and robustness to variations in lesion size, shape, and contrast. Extensive experiments on eight publicly available datasets, including polyp detection, skin lesion segmentation, and ultrasound imaging, show that FM-BFF-Net consistently surpasses recent state-of-the-art methods in Jaccard index and Dice coefficient, confirming its effectiveness and adaptability for diverse medical imaging scenarios.

Problem

• Overcoming limitations in capturing global context in medical image segmentation
• Improving segmentation of structures with complex borders and varying sizes
• Enhancing boundary precision and robustness to lesion variations

Innovation

• Combines convolutional and transformer components for segmentation
• Uses focal modulation attention to enhance context awareness
• Introduces bidirectional feature fusion for cross-scale interactions