Flow-Guided Implicit Neural Representation for Motion-Aware Dynamic MRI Reconstruction

2025-11-20

AI Summary

In dynamic MRI (dMRI), reconstruction artifacts arise from both undersampling and motion corruption. To address this, we propose an end-to-end joint reconstruction framework based on implicit neural representations. Departing from conventional optical flow–guided motion compensation, our approach simultaneously models the dynamic image sequence and a continuous optical flow field via two coupled implicit neural networks—parameterizing spatiotemporal image content and motion, respectively. Crucially, we incorporate physics-informed optical flow constraints—derived from the continuity equation—as a regularizer to enforce motion-aware spatiotemporal consistency. The framework is trained jointly using data consistency loss and differentiable flow regularization, eliminating the need for explicit motion estimation. Evaluated on cardiac dMRI data, our method achieves superior reconstruction quality, higher motion estimation accuracy, and improved temporal fidelity compared to state-of-the-art methods.

Technology Category

• Computer Vision: Motion & Tracking
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Intelligent Robots: Motion and Path Planning

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Dynamic magnetic resonance imaging (dMRI) captures temporally-resolved anatomy but is often challenged by limited sampling and motion-induced artifacts. Conventional motion-compensated reconstructions typically rely on pre-estimated optical flow, which is inaccurate under undersampling and degrades reconstruction quality. In this work, we propose a novel implicit neural representation (INR) framework that jointly models both the dynamic image sequence and its underlying motion field. Specifically, one INR is employed to parameterize the spatiotemporal image content, while another INR represents the optical flow. The two are coupled via the optical flow equation, which serves as a physics-inspired regularization, in addition to a data consistency loss that enforces agreement with k-space measurements. This joint optimization enables simultaneous recovery of temporally coherent images and motion fields without requiring prior flow estimation. Experiments on dynamic cardiac MRI datasets demonstrate that the proposed method outperforms state-of-the-art motion-compensated and deep learning approaches, achieving superior reconstruction quality, accurate motion estimation, and improved temporal fidelity. These results highlight the potential of implicit joint modeling with flow-regularized constraints for advancing dMRI reconstruction.

Problem

• Reconstructs dynamic MRI images from limited k-space sampling data
• Addresses motion artifacts without requiring pre-estimated optical flow
• Simultaneously recovers both image sequences and underlying motion fields

Innovation

• Jointly models dynamic images and motion fields
• Uses implicit neural representation for optical flow
• Couples flow equation with data consistency loss