Spatiotemporal Gaussian representation-based dynamic reconstruction and motion estimation framework for time-resolved volumetric MR imaging (DREME-GSMR)

2026-04-07

AI Summary

This work proposes a sub-second, model-free framework for three-dimensional dynamic MRI reconstruction and motion estimation to meet the real-time imaging demands of motion-adaptive radiotherapy. By unifying anatomical structures and motion fields into 3D Gaussian representations, the method introduces a dual-path MLP/CNN motion encoder that directly infers time-varying motion coefficients from k-space data, complemented by a motion augmentation strategy to enhance robustness against unseen motion patterns. Experiments on XCAT simulations, phantom data, and 26 human subjects demonstrate a temporal resolution of approximately 400 ms and an inference speed of 10 ms per volume. The approach achieves an SSIM of 0.92 on XCAT data with a tumor centroid error as low as 0.50 mm, while yielding centroid errors of 1.31 mm and 0.96 mm in volunteer and patient liver studies, respectively.

Abstract

Time-resolved volumetric MR imaging that reconstructs a 3D MRI within sub-seconds to resolve deformable motion is essential for motion-adaptive radiotherapy. Representing patient anatomy and associated motion fields as 3D Gaussians, we developed a spatiotemporal Gaussian representation-based framework (DREME-GSMR), which enables time-resolved dynamic MRI reconstruction from a pre-treatment 3D MR scan without any prior anatomical/motion model. DREME-GSMR represents a reference MRI volume and a corresponding low-rank motion model (as motion-basis components) using 3D Gaussians, and incorporates a dual-path MLP/CNN motion encoder to estimate temporal motion coefficients of the motion model from raw k-space-derived signals. Furthermore, using the solved motion model, DREME-GSMR can infer motion coefficients directly from new online k-space data, allowing subsequent intra-treatment volumetric MR imaging and motion tracking (real-time imaging). A motion-augmentation strategy is further introduced to improve robustness to unseen motion patterns during real-time imaging. DREME-GSMR was evaluated on the XCAT digital phantom, a physical motion phantom, and MR-LINAC datasets acquired from 6 healthy volunteers and 20 patients (with independent sequential scans for cross-evaluation). DREME-GSMR reconstructs MRIs of a ~400ms temporal resolution, with an inference time of ~10ms/volume. In XCAT experiments, DREME-GSMR achieved mean(s.d.) SSIM, tumor center-of-mass-error(COME), and DSC of 0.92(0.01)/0.91(0.02), 0.50(0.15)/0.65(0.19) mm, and 0.92(0.02)/0.92(0.03) for dynamic reconstruction/real-time imaging. For the physical phantom, the mean target COME was 1.19(0.94)/1.40(1.15) mm for dynamic/real-time imaging, while for volunteers and patients, the mean liver COME for real-time imaging was 1.31(0.82) and 0.96(0.64) mm, respectively.

Problem

• time-resolved volumetric MR imaging
• dynamic reconstruction
• motion estimation
• real-time imaging
• deformable motion

Innovation

• Spatiotemporal Gaussian representation
• Dynamic MRI reconstruction
• Motion estimation
• Real-time imaging
• k-space-based motion tracking