TractoGPT: A GPT architecture for White Matter Segmentation

TL;DR

TractoGPT is a novel GPT-based model designed for automatic white matter bundle segmentation, demonstrating superior accuracy and generalization across multiple datasets by effectively handling structural variability and streamlining data representations.

Contribution

The paper introduces TractoGPT, a GPT architecture tailored for white matter segmentation, combining multiple data representations for improved accuracy and dataset generalization.

Findings

Outperforms state-of-the-art methods on DICE, Overlap, and Overreach scores.

Generalizes well across datasets like TractoInferno and 105HCP.

Retains shape information of white matter bundles effectively.

Abstract

White matter bundle segmentation is crucial for studying brain structural connectivity, neurosurgical planning, and neurological disorders. White Matter Segmentation remains challenging due to structural similarity in streamlines, subject variability, symmetry in 2 hemispheres, etc. To address these challenges, we propose TractoGPT, a GPT-based architecture trained on streamline, cluster, and fusion data representations separately. TractoGPT is a fully-automatic method that generalizes across datasets and retains shape information of the white matter bundles. Experiments also show that TractoGPT outperforms state-of-the-art methods on average DICE, Overlap and Overreach scores. We use TractoInferno and 105HCP datasets and validate generalization across dataset.