A foundation for systematic analysis of transformers and RNNs for tractography
Abstract
Machine learning (ML) has emerged as a promising approach for improving diffusion MRI (dMRI) tractography, a task that remains limited by the intrinsic tension between local diffusion information and global anatomical plausibility. In this work, we systematically evaluate recurrent neural networks (RNNs) and Transformer models for iterative tractography, with particular attention to training strategies, input representations (including convolutional neural network (CNN)-based embeddings and end-of-sequence (EOS) tokens), and hyperparameter selection. We introduce a generation-validation phase enabling supervision at the streamline level during training, allowing supervision despite the mismatch between local loss functions and global streamline quality. Using the ISMRM2015 tractography challenge dataset, our models achieve the highest reported performance to date. Through controlled experiments, we quantify the impact of missing bundles, noisy or imperfect training streamlines, and invalid fibers in the training set. Finally, we demonstrate the applicability of our best-performing models for in vivo data from the Tractoinferno database. Overall, our results highlight both the potential and the limits of sequence-based deep learning models such as Transformers and RNNs for tractography, and emphasize the need for improved phantoms and evaluation methods for in vivo validation. We provide takeaways and recommendations for future researchers training and validating sequence-based supervised methods for tractography.
Related
- Transformers for fMRI analysis: from spatiotemporal modeling to brain foundation models.
- Learning reveals invisible structure in low-rank RNNs
- Prediction emerges in RNNs trained for perception
- Oscillatory Phase Coding Emerges in Adaptive RNNS for Flexible Working Memory
- Oscillatory Phase Coding Emerges in Adaptive RNNs for Flexible Working Memory
BibTeX
@misc{renauld2026foundation,
title = {{A foundation for systematic analysis of transformers and RNNs for tractography}},
author = {Emmanuelle Renauld and Philippe Poulin and Hugo Larochelle and Antoine Théberge and Maxime Descoteaux},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.01894},
eprint = {2610.01894},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.01894},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - A foundation for systematic analysis of transformers and RNNs for tractography
AU - Emmanuelle Renauld
AU - Philippe Poulin
AU - Hugo Larochelle
AU - Antoine Théberge
AU - Maxime Descoteaux
PY - 2026
DA - 2026/10/01
PB - arXiv
DO - 10.48550/arxiv.2610.01894
UR - https://arxiv.org/abs/2610.01894
AB - Machine learning (ML) has emerged as a promising approach for improving diffusion MRI (dMRI) tractography, a task that remains limited by the intrinsic tension between local diffusion information and global anatomical plausibility. In this work, we systematically evaluate recurrent neural networks (RNNs) and Transformer models for iterative tractography, with particular attention to training strategies, input representations (including convolutional neural network (CNN)-based embeddings and end-of-sequence (EOS) tokens), and hyperparameter selection. We introduce a generation-validation phase enabling supervision at the streamline level during training, allowing supervision despite the mismatch between local loss functions and global streamline quality. Using the ISMRM2015 tractography challenge dataset, our models achieve the highest reported performance to date. Through controlled experiments, we quantify the impact of missing bundles, noisy or imperfect training streamlines, and invalid fibers in the training set. Finally, we demonstrate the applicability of our best-performing models for in vivo data from the Tractoinferno database. Overall, our results highlight both the potential and the limits of sequence-based deep learning models such as Transformers and RNNs for tractography, and emphasize the need for improved phantoms and evaluation methods for in vivo validation. We provide takeaways and recommendations for future researchers training and validating sequence-based supervised methods for tractography.
ER -