Leveraging rotational equivariance for reinforcement learning in tractography.

Fabian Leander Sinzinger¹, Antoine Théberge², Pierre-Marc Jodoin², Maxime Descoteaux³, Rodrigo Moreno⁴

¹KTH Royal Institute of Technology, Department of Biomedical Engineering and Health Systems, Hälsovägen 11C, Huddinge, 14157, Stockholm, Sweden. Electronic address: fabiansi@kth.se.

²Videos and Images Theory and Analytics Laboratory (VITAL), Faculté des Sciences, Université de Sherbrooke, 2500, boul. de l'Université, Sherbrooke, J1K2R1, Québec, Canada.

³Sherbrooke Connectivity Imaging Laboratory (SCIL), Department of Computer Science, Université de Sherbrooke, 2500, boul. de l'Université, Sherbrooke, J1K2R1, Québec, Canada.

⁴KTH Royal Institute of Technology, Department of Biomedical Engineering and Health Systems, Hälsovägen 11C, Huddinge, 14157, Stockholm, Sweden; Department of Neurobiology, Care Sciences and Society, Karolinska Institutet, Blickagången 16, 14183 Huddinge, Sweden. Electronic address: rodmore@kth.se.

PubMed DOI

Abstract

Brain tractography involves mapping diffusion-weighted images (DWI) onto streamlines representing neural fibre bundles. Recent research avenues have framed tractography into a reinforcement learning (RL) framework with actor-critic models. However, previous RL-based methods may compromise geometrical relations between the input (DWI) and output (tractogram). More specifically, 3D rotations applied…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free Account Sign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.