Reliable estimation of the number of compartments in diffusion MRI

Koppers S, Haarburger C, Edgar JC, Merhof D (2017)

Publication Type: Conference contribution

Publication year: 2017

Journal

Informatik aktuell Gesellschaft für Informatik (GI)

Publisher: Kluwer Academic Publishers

Pages Range: 203-208

Conference Proceedings Title: Informatik aktuell

Event location: Heidelberg, DEU

ISBN: 9783662543443

DOI: 10.1007/978-3-662-54345-0_46

Abstract

A-priori knowledge of the number of fibers in a voxel is mandatory and crucial when reconstructing multi-fiber voxels in diffusion MRI. Especially for clinical purposes, this estimation needs to be stable, even when only few gradient directions are acquired. In this work, we propose a novel approach to address this problem based on a deep convolutional neural network (CNN), which is able to identify important gradient directions and can be directly trained on real data. To obtain a ground truth using real data, 100 uncorrelated Human Connectome Project datasets are utilized, with a state-of-the-art framework used for generating a relative ground truth. It is shown that this CNN approach outperforms other state-of-the-art machine learning approaches.

Involved external institutions

Rheinisch-Westfälische Technische Hochschule (RWTH) Aachen DE Germany (DE) University of Pennsylvania US United States (USA) (US)

How to cite

APA:

Koppers, S., Haarburger, C., Edgar, J.C., & Merhof, D. (2017). Reliable estimation of the number of compartments in diffusion MRI. In Klaus Hermann Maier-Hein, Heinz Handels, Thomas Martin Deserno, Thomas Tolxdorff (Eds.), Informatik aktuell (pp. 203-208). Heidelberg, DEU: Kluwer Academic Publishers.

MLA:

Koppers, Simon, et al. "Reliable estimation of the number of compartments in diffusion MRI." Proceedings of the Workshops on Image processing for the medicine, 2017, Heidelberg, DEU Ed. Klaus Hermann Maier-Hein, Heinz Handels, Thomas Martin Deserno, Thomas Tolxdorff, Kluwer Academic Publishers, 2017. 203-208.

BibTeX: Download