World of Forms: Deformable Geometric Templates for One-Shot Surface Meshing in Coronary CT Angiography

TL;DR

This paper introduces a data-efficient deep learning method using geometric priors and graph neural networks for direct 3D surface meshing in coronary CT angiography, improving accuracy and topological consistency with less data.

Contribution

It presents a novel deformable geometric template approach with a multi-resolution graph neural network and a masked autoencoder pretraining strategy for improved 3D surface meshing.

Findings

Outperforms nnUNet in one-shot segmentation tasks.

Produces mesh quality comparable or superior to marching cubes.

Flexible template choice enhances topological accuracy.

Abstract

Deep learning-based medical image segmentation and surface mesh generation typically involve a sequential pipeline from image to segmentation to meshes, often requiring large training datasets while making limited use of prior geometric knowledge. This may lead to topological inconsistencies and suboptimal performance in low-data regimes. To address these challenges, we propose a data-efficient deep learning method for direct 3D anatomical object surface meshing using geometric priors. Our approach employs a multi-resolution graph neural network that operates on a prior geometric template which is deformed to fit object boundaries of interest. We show how different templates may be used for the different surface meshing targets, and introduce a novel masked autoencoder pretraining strategy for 3D spherical data. The proposed method outperforms nnUNet in a one-shot setting for segmentation of the pericardium, left ventricle (LV) cavity and the LV myocardium. Similarly, the method outperforms other lumen segmentation operating on multi-planar reformatted images. Results further indicate that mesh quality is on par with or improves upon marching cubes post-processing of voxel mask predictions, while remaining flexible in the choice of mesh triangulation prior, thus paving the way for more accurate and topologically consistent 3D medical object surface meshing.