Wavelet-Optimized Motion Artifact Correction in 3D MRI Using Pre-trained 2D Score Priors

2025-11-04

AI Summary

MRI motion artifacts severely degrade image quality and diagnostic reliability. Existing 3D score-based generative modeling (SGM) approaches require explicit knowledge of the forward operator and suffer from slow inference, hindering clinical deployment. To address these limitations, we propose the first forward-model-free, end-to-end 3D MRI motion artifact correction framework. Our method leverages a pre-trained 2D score prior to guide 3D distribution modeling, integrates mean-regression stochastic differential equations with a novel Fourier-domain diffusion mechanism (“Fourier Diffusion”), and employs Fourier convolution for accelerated feature extraction and inference. Evaluated on both simulated and real clinical datasets, our approach significantly outperforms state-of-the-art methods—achieving PSNR and SSIM improvements exceeding 2.1 dB and 0.03, respectively, while accelerating inference by 5.8×. The framework delivers high reconstruction fidelity, strong robustness to diverse motion patterns, and practical clinical applicability.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Motion artifacts in magnetic resonance imaging (MRI) remain a major challenge, as they degrade image quality and compromise diagnostic reliability. Score-based generative models (SGMs) have recently shown promise for artifact removal. However, existing 3D SGM-based approaches are limited in two key aspects: (1) their strong dependence on known forward operators makes them ineffective for correcting MRI motion artifacts, and (2) their slow inference speed hinders clinical translation. To overcome these challenges, we propose a wavelet-optimized end-to-end framework for 3D MRI motion correct using pre-trained 2D score priors (3D-WMoCo). Specifically, two orthogonal 2D score priors are leveraged to guide the 3D distribution prior, while a mean-reverting stochastic differential equation (SDE) is employed to model the restoration process of motion-corrupted 3D volumes to motion-free 3D distribution. Furthermore, wavelet diffusion is introduced to accelerate inference, and wavelet convolution is applied to enhance feature extraction. We validate the effectiveness of our approach through both simulated motion artifact experiments and real-world clinical motion artifact correction tests. The proposed method achieves robust performance improvements over existing techniques. Implementation details and source code are available at: https://github.com/ZG-yuan/3D-WMoCo.

Problem

• Correcting motion artifacts in 3D MRI scans
• Overcoming slow inference speed in artifact removal
• Using 2D score priors for 3D motion correction

Innovation

• Using pre-trained 2D score priors for 3D MRI correction
• Employing wavelet diffusion to accelerate inference speed
• Applying wavelet convolution to enhance feature extraction