White Matter Tract Segmentation is imperative for studying brain structural connectivity, neurological disorders and neurosurgery. This task remains complex, as tracts differ among themselves, across subjects and conditions, yet have similar 3D structure across hemispheres and subjects. To address these challenges, we propose TrackletGPT, a language-like GPT framework which reintroduces sequential information in tokens using tracklets. TrackletGPT generalises seamlessly across datasets, is fully automatic, and encodes granular sub-streamline segments, Tracklets, scaling and refining GPT models in Tractography Segmentation. Based on our experiments, TrackletGPT outperforms state-of-the-art methods on average DICE, Overlap and Overreach scores on TractoInferno and HCP datasets, even on inter-dataset experiments.
@article{arxiv.2601.13935,
title = {TrackletGPT: A Language-like GPT Framework for White Matter Tract Segmentation},
author = {Anoushkrit Goel and Simroop Singh and Ankita Joshi and Ranjeet Ranjan Jha and Chirag Ahuja and Aditya Nigam and Arnav Bhavsar},
journal= {arXiv preprint arXiv:2601.13935},
year = {2026}
}
Comments
Accepted at 23rd IEEE International Symposium on Biomedical Imaging (ISBI), 2026