Weakly Supervised Learning of Cortical Surface Reconstruction from Segmentations

TL;DR

This paper introduces CoSeg, a weakly supervised learning framework for cortical surface reconstruction from brain segmentations, eliminating the need for traditional pseudo ground truth surfaces and enabling fast, accurate results on adult and fetal MRI.

Contribution

CoSeg is a novel weakly supervised approach that uses temporal attention networks to learn cortical surface deformations directly from segmentations, bypassing traditional pGT surface generation.

Findings

Achieves superior geometric accuracy over existing methods.

Reconstructs cortical surfaces from fetal MRI where traditional pipelines fail.

Fast processing time of 0.11 seconds per hemisphere.

Abstract

Existing learning-based cortical surface reconstruction approaches heavily rely on the supervision of pseudo ground truth (pGT) cortical surfaces for training. Such pGT surfaces are generated by traditional neuroimage processing pipelines, which are time consuming and difficult to generalize well to low-resolution brain MRI, e.g., from fetuses and neonates. In this work, we present CoSeg, a learning-based cortical surface reconstruction framework weakly supervised by brain segmentations without the need for pGT surfaces. CoSeg introduces temporal attention networks to learn time-varying velocity fields from brain MRI for diffeomorphic surface deformations, which fit an initial surface to target cortical surfaces within only 0.11 seconds for each brain hemisphere. A weakly supervised loss is designed to reconstruct pial surfaces by inflating the white surface along the normal direction towards the boundary of the cortical gray matter segmentation. This alleviates partial volume effects and encourages the pial surface to deform into deep and challenging cortical sulci. We evaluate CoSeg on 1,113 adult brain MRI at 1mm and 2mm resolution. CoSeg achieves superior geometric and morphological accuracy compared to existing learning-based approaches. We also verify that CoSeg can extract high-quality cortical surfaces from fetal brain MRI on which traditional pipelines fail to produce acceptable results.