Physics informed guided diffusion for accelerated multi-parametric MRI reconstruction

2025-06-29

AI Summary

To address low measurement fidelity and physical model inconsistency in multiparametric mapping reconstruction for highly accelerated, transient quantitative MRI (e.g., magnetic resonance fingerprinting, MRF), this paper proposes a physics-guided diffusion model reconstruction framework. The method integrates a pre-trained denoising diffusion prior into a proximal splitting optimization scheme, jointly enforcing k-space data consistency and Bloch equation response constraints to achieve stable and accurate inverse problem solving. Compared with conventional compressed sensing and end-to-end deep learning approaches, our method significantly improves the accuracy of quantitative parameter maps (e.g., T₁ and T₂) on in vivo human brain data, while rigorously preserving measurement fidelity and ensuring biophysical interpretability. By unifying data-driven priors with first-principles physical modeling, the proposed framework establishes a new paradigm for rapid, reliable, and verifiable quantitative MRI.

Technology Category

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

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Security and Privacy: Data transparency and provenance

Abstract

We introduce MRF-DiPh, a novel physics informed denoising diffusion approach for multiparametric tissue mapping from highly accelerated, transient-state quantitative MRI acquisitions like Magnetic Resonance Fingerprinting (MRF). Our method is derived from a proximal splitting formulation, incorporating a pretrained denoising diffusion model as an effective image prior to regularize the MRF inverse problem. Further, during reconstruction it simultaneously enforces two key physical constraints: (1) k-space measurement consistency and (2) adherence to the Bloch response model. Numerical experiments on in-vivo brain scans data show that MRF-DiPh outperforms deep learning and compressed sensing MRF baselines, providing more accurate parameter maps while better preserving measurement fidelity and physical model consistency-critical for solving reliably inverse problems in medical imaging.

Problem

• Accelerated multi-parametric MRI reconstruction using physics-informed diffusion
• Enforcing k-space consistency and Bloch model adherence in MRF
• Improving accuracy and fidelity in quantitative tissue mapping

Innovation

• Physics informed denoising diffusion model
• Enforces k-space measurement consistency
• Adheres to Bloch response model