CA-Diff: Collaborative Anatomy Diffusion for Brain Tissue Segmentation

2025-06-28

AI Summary

Existing CNN- and Transformer-based methods struggle to capture complex anatomical structures in brain MRI segmentation, while current diffusion models neglect anatomical priors, limiting their performance. To address these issues, this paper proposes the Collaborative Anatomical Diffusion Framework (CADF). CADF encodes global anatomical spatial constraints via distance field representations and introduces a consistency loss alongside a time-adaptive channel attention module to jointly model anatomical priors and image features. Within a U-Net backbone, CADF integrates conditional guided diffusion with time-varying attention to enhance structural boundary accuracy. Evaluated on multiple public benchmarks, CADF consistently outperforms state-of-the-art methods, achieving an average Dice score improvement of 2.1%. Notably, it demonstrates superior robustness for small structures and lesion regions. CADF establishes a novel, interpretable, and anatomy-constrained paradigm for generative-model-based medical image segmentation.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Segmentation of brain structures from MRI is crucial for evaluating brain morphology, yet existing CNN and transformer-based methods struggle to delineate complex structures accurately. While current diffusion models have shown promise in image segmentation, they are inadequate when applied directly to brain MRI due to neglecting anatomical information. To address this, we propose Collaborative Anatomy Diffusion (CA-Diff), a framework integrating spatial anatomical features to enhance segmentation accuracy of the diffusion model. Specifically, we introduce distance field as an auxiliary anatomical condition to provide global spatial context, alongside a collaborative diffusion process to model its joint distribution with anatomical structures, enabling effective utilization of anatomical features for segmentation. Furthermore, we introduce a consistency loss to refine relationships between the distance field and anatomical structures and design a time adapted channel attention module to enhance the U-Net feature fusion procedure. Extensive experiments show that CA-Diff outperforms state-of-the-art (SOTA) methods.

Problem

• Improves brain MRI segmentation accuracy using anatomical features
• Integrates distance field for global spatial context in segmentation
• Enhances diffusion models with collaborative anatomy and attention modules

Innovation

• Integrates spatial anatomical features for segmentation
• Uses distance field as anatomical condition
• Introduces consistency loss and attention module