NC-Reg : Neural Cortical Maps for Rigid Registration

TL;DR

This paper introduces neural cortical maps as a continuous neural representation for cortical features, enabling faster and more accurate rigid registration of cortical surfaces, with potential clinical applications.

Contribution

It proposes neural cortical maps as a novel, efficient representation for cortical features and introduces NC-Reg, a new algorithm for rigid registration using these maps.

Findings

Achieves up to 30x faster runtimes than traditional methods.

Demonstrates sub-degree registration accuracy (<1°).

Serves as a robust pre-alignment strategy in clinical settings.

Abstract

We introduce neural cortical maps, a continuous and compact neural representation for cortical feature maps, as an alternative to traditional discrete structures such as grids and meshes. It can learn from meshes of arbitrary size and provide learnt features at any resolution. Neural cortical maps enable efficient optimization on the sphere and achieve runtimes up to 30 times faster than classic barycentric interpolation (for the same number of iterations). As a proof of concept, we investigate rigid registration of cortical surfaces and propose NC-Reg, a novel iterative algorithm that involves the use of neural cortical feature maps, gradient descent optimization and a simulated annealing strategy. Through ablation studies and subject-to-template experiments, our method demonstrates sub-degree accuracy ($<1^\circ$ from the global optimum), and serves as a promising robust pre-alignment strategy, which is critical in clinical settings.