Multi-Contrast MRI Motion Correction via Parameter-Informed Disentanglement and Adaptive Experts

2026-05-28

AI Summary

Motion artifacts in multi-contrast MRI significantly compromise diagnostic reliability, yet existing deep learning methods exhibit limited generalization across modalities and varying artifact severities. This work proposes a unified framework that leverages the ScanCLIP pre-trained model to generate contrast embeddings from scan parameters, effectively disentangling anatomical content from contrast-specific style. Furthermore, a Mixture-of-Experts network—integrating a Vision Transformer with a dual-path decoder—is introduced to enable severity-adaptive artifact correction. The proposed method achieves a PSNR gain of 0.75 dB and an SSIM improvement of up to 0.0279 on the IXI and HCP datasets, while demonstrating strong zero-shot generalization on unseen real-world clinical data.

Abstract

Motion artifacts in magnetic resonance imaging (MRI) degrade diagnostic reliability. Existing deep learning methods are typically contrast-specific and fail to generalize across diverse modalities and artifact severities. We propose a unified framework combining parameter-informed contrast disentanglement with severity-aware adaptive correction. ScanCLIP, pretrained on over 30,000 MRI text-image pairs, derives contrast embeddings from acquisition parameters to disentangle contrast style from anatomical content, yielding contrast-free features. A Vision Transformer then estimates motion severity and routes features through a Mixture-of-Experts network, enabling targeted artifact correction. A dual-pathway decoder reconstructs both the clean image and residual artifact map, enforcing image-space consistency. On IXI and HCP benchmarks, our method improves PSNR by 0.75 dB and SSIM by up to 0.0279 over state-of-the-art approaches, with larger gains at higher artifact severities. It further demonstrates robust zero-shot generalization on real-world clinical data acquired with unseen scanning parameters, where existing methods either fail to remove artifacts or introduce additional distortions.

Problem

• Motion artifacts
• Multi-contrast MRI
• Generalization
• Deep learning
• Diagnostic reliability

Innovation

• contrast disentanglement
• adaptive experts
• motion correction
• zero-shot generalization
• Mixture-of-Experts