Antoine Théberge

The paper develops an automated pipeline for segmenting and parcellating brainstem and subcortical white matter pathways using diffusion MRI data.

The paper evaluates the reliability and variability of diffusion- and myelin-based MRI measurements across multiple sessions and fiber populations, finding moderate to high reliability for diffusion measures and bundle-dependent variability.

Track-To-Learn uses deep reinforcement learning to improve diffusion MRI tractography without requiring labeled data, achieving better generalization than prior methods.