A Deep-Discrete Learning Framework for Spherical Surface Registration

TL;DR

This paper introduces an unsupervised deep learning framework for spherical surface registration that improves alignment accuracy and deformation smoothness over classical methods, using a geometric deep learning approach with CRF regularization.

Contribution

It presents a novel spherical geometric deep learning framework converting registration into a multi-label classification problem, with end-to-end training and regularization, outperforming existing methods.

Findings

Performs competitively with classical algorithms in similarity and distortion.

Produces smoother deformations than other learning-based methods.

Effective even on subjects with atypical cortical morphology.

Abstract

Cortical surface registration is a fundamental tool for neuroimaging analysis that has been shown to improve the alignment of functional regions relative to volumetric approaches. Classically, image registration is performed by optimizing a complex objective similarity function, leading to long run times. This contributes to a convention for aligning all data to a global average reference frame that poorly reflects the underlying cortical heterogeneity. In this paper, we propose a novel unsupervised learning-based framework that converts registration to a multi-label classification problem, where each point in a low-resolution control grid deforms to one of fixed, finite number of endpoints. This is learned using a spherical geometric deep learning architecture, in an end-to-end unsupervised way, with regularization imposed using a deep Conditional Random Field (CRF). Experiments show that our proposed framework performs competitively, in terms of similarity and areal distortion, relative to the most popular classical surface registration algorithms and generates smoother deformations than other learning-based surface registration methods, even in subjects with atypical cortical morphology.