Live-view 4D GRASP MRI: A framework for robust real-time respiratory motion tracking with a sub-second imaging...

Area of Science:

• Medical Imaging
• Magnetic Resonance Imaging
• Radiotherapy Technology

Background:

• Real-time volumetric MRI is crucial for dynamic imaging and motion-adaptive radiotherapy.
• Existing methods often face latency or fidelity challenges in capturing volumetric data in real-time.
• Golden-angle radial sparse parallel (GRASP) MRI offers a promising approach for accelerated volumetric acquisition.

Purpose of the Study:

• To propose and evaluate a novel live-view GRASP MRI framework for low-latency, high-fidelity real-time volumetric MRI.
• To assess the accuracy and robustness of the proposed framework, particularly the role of 2D navigators in motion characterization.

Main Methods:

• The live-view GRASP MRI framework employs a two-stage approach: an off-view stage for data acquisition and database generation, and a live-view stage for real-time image reconstruction.
• A navi-stack-of-stars sampling scheme acquires 3D k-space data and 2D navigators in the off-view stage, creating a 4D motion database.
• In the live-view stage, 2D navigators guide the selection of pre-acquired 3D data for real-time volumetric image generation.

Main Results:

• The live-view GRASP MRI framework successfully generates real-time volumetric images with high fidelity, closely matching ground-truth references.
• An imaging latency below 500 milliseconds was achieved, enabling truly real-time imaging.
• The use of 2D navigators demonstrated superior robustness and accuracy in characterizing respiratory and body movements compared to 1D navigators.

Conclusions:

• Live-view GRASP MRI provides a novel, accurate, and robust solution for real-time volumetric imaging.
• The framework's low latency and high fidelity make it suitable for motion-adaptive radiotherapy, particularly on MRI-Linac systems.