Predicting 4D Liver MRI for MR-guided Interventions

TL;DR

This paper introduces a deep learning approach to generate real-time, high-resolution 4D liver MRI from 2D navigator slices, improving imaging speed and quality for MR-guided interventions and motion analysis.

Contribution

The study presents a novel CNN-based method for real-time 4D MRI prediction from 2D slices, enabling high-resolution imaging with reduced acquisition times for interventions.

Findings

Achieved mean target registration error of 1.19mm, below voxel size.

Produced high-quality 4D MRI with 0.6s/volume temporal resolution.

Small training datasets (2-24 min acquisition) suffice for promising results.

Abstract

Organ motion poses an unresolved challenge in image-guided interventions. In the pursuit of solving this problem, the research field of time-resolved volumetric magnetic resonance imaging (4D MRI) has evolved. However, current techniques are unsuitable for most interventional settings because they lack sufficient temporal and/or spatial resolution or have long acquisition times. In this work, we propose a novel approach for real-time, high-resolution 4D MRI with large fields of view for MR-guided interventions. To this end, we trained a convolutional neural network (CNN) end-to-end to predict a 3D liver MRI that correctly predicts the liver's respiratory state from a live 2D navigator MRI of a subject. Our method can be used in two ways: First, it can reconstruct near real-time 4D MRI with high quality and high resolution (209x128x128 matrix size with isotropic 1.8mm voxel size and 0.6s/volume) given a dynamic interventional 2D navigator slice for guidance during an intervention. Second, it can be used for retrospective 4D reconstruction with a temporal resolution of below 0.2s/volume for motion analysis and use in radiation therapy. We report a mean target registration error (TRE) of 1.19 $\pm$0.74mm, which is below voxel size. We compare our results with a state-of-the-art retrospective 4D MRI reconstruction. Visual evaluation shows comparable quality. We show that small training sizes with short acquisition times down to 2min can already achieve promising results and 24min are sufficient for high quality results. Because our method can be readily combined with earlier methods, acquisition time can be further decreased while also limiting quality loss. We show that an end-to-end, deep learning formulation is highly promising for 4D MRI reconstruction.