SupWMA: Consistent and Efficient Tractography Parcellation of Superficial White Matter with Deep Learning

TL;DR

SupWMA is a deep learning framework that efficiently and accurately parcellates superficial white matter from whole-brain tractography, outperforming existing methods in consistency, speed, and robustness across diverse datasets.

Contribution

The paper introduces a novel deep learning approach with supervised contrastive learning for SWM parcellation, addressing a complex area less explored by prior methods.

Findings

SupWMA achieves high consistency and accuracy in SWM parcellation.

SupWMA is significantly faster than existing methods.

The method generalizes well across diverse datasets.

Abstract

White matter parcellation classifies tractography streamlines into clusters or anatomically meaningful tracts to enable quantification and visualization. Most parcellation methods focus on the deep white matter (DWM), while fewer methods address the superficial white matter (SWM) due to its complexity. We propose a deep-learning-based framework, Superficial White Matter Analysis (SupWMA), that performs an efficient and consistent parcellation of 198 SWM clusters from whole-brain tractography. A point-cloud-based network is modified for our SWM parcellation task, and supervised contrastive learning enables more discriminative representations between plausible streamlines and outliers. We perform evaluation on a large tractography dataset with ground truth labels and on three independently acquired testing datasets from individuals across ages and health conditions. Compared to several state-of-the-art methods, SupWMA obtains a highly consistent and accurate SWM parcellation result. In addition, the computational speed of SupWMA is much faster than other methods.