FOD-Diff: 3D Multi-Channel Patch Diffusion Model for Fiber Orientation Distribution

2025-12-17

AI Summary

To address the trade-off between low angular resolution and poor fiber orientation distribution (FOD) reconstruction accuracy in single-shell low-angular-resolution diffusion MRI (LAR-dMRI), and the prohibitively long acquisition time of multi-shell high-angular-resolution FOD (HAR-FOD) in clinical practice, this work proposes a novel generative framework. Methodologically: (i) we introduce the first FOD-patch adapter that incorporates anatomical priors to guide local FOD modeling; (ii) we design a voxel-wise conditional coordination module to enforce global structural consistency; and (iii) we propose a spherical harmonic (SH) attention mechanism to explicitly capture nonlinear interdependencies among SH coefficients. Built upon a 3D multi-channel patch-based diffusion architecture, our method significantly improves angular resolution fidelity and SH coefficient reconstruction accuracy for HAR-FOD. It maintains computational efficiency and demonstrates strong potential for clinical deployment, outperforming state-of-the-art methods in comprehensive evaluation.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Natural Language Processing: Generation
• Machine Learning: Learning with Manifolds

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Diffusion MRI (dMRI) is a critical non-invasive technique to estimate fiber orientation distribution (FOD) for characterizing white matter integrity. Estimating FOD from single-shell low angular resolution dMRI (LAR-FOD) is limited by accuracy, whereas estimating FOD from multi-shell high angular resolution dMRI (HAR-FOD) requires a long scanning time, which limits its applicability. Diffusion models have shown promise in estimating HAR-FOD based on LAR-FOD. However, using diffusion models to efficiently generate HAR-FOD is challenging due to the large number of spherical harmonic (SH) coefficients in FOD. Here, we propose a 3D multi-channel patch diffusion model to predict HAR-FOD from LAR-FOD. We design the FOD-patch adapter by introducing the prior brain anatomy for more efficient patch-based learning. Furthermore, we introduce a voxel-level conditional coordinating module to enhance the global understanding of the model. We design the SH attention module to effectively learn the complex correlations of the SH coefficients. Our experimental results show that our method achieves the best performance in HAR-FOD prediction and outperforms other state-of-the-art methods.

Problem

• Estimates high-resolution fiber orientation distribution from low-resolution MRI
• Reduces scanning time for accurate white matter characterization
• Models complex spherical harmonic coefficients efficiently using diffusion

Innovation

• 3D multi-channel patch diffusion model for FOD prediction
• FOD-patch adapter with prior brain anatomy for efficient learning
• SH attention module to learn complex spherical harmonic correlations