Mesh-based 3D Motion Tracking in Cardiac MRI using Deep Learning

TL;DR

This paper introduces a deep learning method that models the heart as a 3D mesh and estimates its motion from 2D cardiac MRI images, improving accuracy and correspondence for cardiac function assessment.

Contribution

The novel approach uses a differentiable mesh rasterizer to enable end-to-end training and accurate 3D motion estimation directly from 2D multi-view MRI images.

Findings

Outperforms conventional motion tracking methods.

Maintains vertex correspondence over time.

Validated on UK Biobank data.

Abstract

3D motion estimation from cine cardiac magnetic resonance (CMR) images is important for the assessment of cardiac function and diagnosis of cardiovascular diseases. Most of the previous methods focus on estimating pixel-/voxel-wise motion fields in the full image space, which ignore the fact that motion estimation is mainly relevant and useful within the object of interest, e.g., the heart. In this work, we model the heart as a 3D geometric mesh and propose a novel deep learning-based method that can estimate 3D motion of the heart mesh from 2D short- and long-axis CMR images. By developing a differentiable mesh-to-image rasterizer, the method is able to leverage the anatomical shape information from 2D multi-view CMR images for 3D motion estimation. The differentiability of the rasterizer enables us to train the method end-to-end. One advantage of the proposed method is that by tracking the motion of each vertex, it is able to keep the vertex correspondence of 3D meshes between time frames, which is important for quantitative assessment of the cardiac function on the mesh. We evaluate the proposed method on CMR images acquired from the UK Biobank study. Experimental results show that the proposed method quantitatively and qualitatively outperforms both conventional and learning-based cardiac motion tracking methods.