Motion-compensated cardiac MRI using low-rank diffeomorphic flow (DMoCo)

2025-05-06

AI Summary

To address respiratory and cardiac motion artifacts in free-breathing, non-gated 3D cardiac MRI, this paper proposes an unsupervised motion-compensated reconstruction method. The approach jointly optimizes a static template image and a parametric motion model directly from k-space data—without external motion monitoring or labeled ground truth. Its key contributions are: (i) the first low-rank structured parametric velocity field model, which compactly represents the full-cycle diffeomorphic deformation flow; and (ii) an end-to-end differentiable optimization framework integrating variational motion estimation with low-rank tensor decomposition, ensuring physically interpretable motion modeling. By unifying differentiable diffeomorphic integration and spectral regularization, the method achieves superior motion resolution and compensation compared to state-of-the-art techniques. Quantitative and qualitative evaluations on free-breathing 3D cine MRI demonstrate significant improvements in spatial resolution, temporal consistency, and artifact suppression.

Technology Category

• Computer Vision: Motion & Tracking
• Machine Learning: Matrix & Tensor Methods
• Intelligent Robots: Motion and Path Planning

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
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

We introduce an unsupervised motion-compensated image reconstruction algorithm for free-breathing and ungated 3D cardiac magnetic resonance imaging (MRI). We express the image volume corresponding to each specific motion phase as the deformation of a single static image template. The main contribution of the work is the low-rank model for the compact joint representation of the family of diffeomorphisms, parameterized by the motion phases. The diffeomorphism at a specific motion phase is obtained by integrating a parametric velocity field along a path connecting the reference template phase to the motion phase. The velocity field at different phases is represented using a low-rank model. The static template and the low-rank motion model parameters are learned directly from the k-space data in an unsupervised fashion. The more constrained motion model is observed to offer improved recovery compared to current motion-resolved and motion-compensated algorithms for free-breathing 3D cine MRI.

Problem

• Unsupervised motion compensation in free-breathing cardiac MRI
• Low-rank model for compact diffeomorphic flow representation
• Improved image recovery compared to current motion-resolved methods

Innovation

• Unsupervised motion-compensated reconstruction algorithm
• Low-rank model for diffeomorphisms representation
• Parametric velocity field integration for deformation