Prospective Dynamic 3D MRI Reconstruction via Latent-Space Motion Tracking from Single Measurement

2026-06-02

AI Summary

This work addresses the challenge of balancing image fidelity and motion estimation efficiency in prospective dynamic 3D MRI reconstruction under ultra-sparse sampling and low-latency constraints. The authors propose the PDMR framework, which uniquely integrates a low-dimensional latent manifold model of motion fields with a triplane geometric representation. By learning motion priors offline and leveraging triplanes for efficient online encoding of deformation vector fields, PDMR substantially reduces the optimization search space. The method enables real-time reconstruction of high-fidelity, temporally consistent dynamic MRI sequences from a single measurement. Evaluations on both XCAT phantoms and in vivo abdominal data demonstrate that PDMR outperforms existing retrospective and online approaches in prospective scenarios, including Immediate and After-2min acquisition protocols.

Abstract

Prospective reconstruction is crucial in many clinical applications such as MRI-guided radiotherapy, which demands accurate image reconstruction and fast motion estimation from currently acquired measurements. However, prospective reconstruction remains challenging due to ultra-sparse sampling and stringent latency requirements. In this work, we propose PDMR, a Prospective Dynamic 3D MRI Reconstruction framework with latent-space motion tracking. Our core idea is to learn an efficient and generalizable latent manifold of motion fields offline, enabling rapid online adaptation for prospective reconstruction. Specifically, we parameterize the deformation vector fields (DVFs) on a low-dimensional manifold, effectively reducing the search space for fast online adaptation, and employ a tri-plane representation to achieve geometry-aware and memory-efficient encoding of 3D motion. Experiments on both XCAT digital phantoms and in-house abdominal MRI datasets demonstrate that PDMR achieves high-fidelity and temporally consistent reconstruction across multiple prospective scenarios (Immediate and After-2min), outperforming state-of-the-art retrospective and online methods. Our results suggest a promising pathway toward ultra-fast, motion-aware prospective MRI reconstruction in clinical practice.

Problem

• prospective reconstruction
• dynamic MRI
• motion estimation
• ultra-sparse sampling
• low-latency imaging

Innovation

• prospective MRI reconstruction
• latent-space motion tracking
• deformation vector fields
• tri-plane representation
• dynamic 3D MRI