Abstract #0852

TractLearn: a geodesic learning framework for quantitative dissection of brain bundles

Arnaud Attyé^1,2, Felix Renard³, Monica Baciu², Elise Roger², Laurent Lamalle⁴, Patrick Dehail⁵, Hélène Cassoudesalle⁵, and Fernando Calamante^6,7

¹School of Biomedical Engineering, University of Sydney, Sydney, Australia, ²CNRS LPNC UMR 5105, University of Grenoble Alpes, Grenoble, France, ³Laboratoire d'informatique de Grenoble, Grenoble, France, ⁴University of Grenoble Alpes, Grenoble, France, ⁵Bordeaux Universitary Hospital, Bordeaux, France, ⁶Sydney Imaging Lab, University of Sydney, Sydney, Australia, ⁷School of Aerospace, Mechanical and Mechatronic Engineering, Sydney, Australia

Here we present a unified framework for brain fascicles quantitative analyses by geodesic learning (TractLearn) — as a data-driven unsupervised learning task. TractLearn allows a mapping between the image high-dimensional domain and the reduced latent space of brain fascicles. Besides providing a framework to test the reliability of various brain metrics with a global overview, it allows to identify subtle quantitative alteration in disease model with small subset of patients and/or data sparsity. With this regard, TractLearn is a ready-to-use algorithm for precision medicine.

This abstract and the presentation materials are available to members only; a login is required.

Join Here