Meeting Banner
Abstract #4475

Clinical Recovery of intracellular volume fraction and fiberODF for a patient with asymptomatic temporal-occipital lesion using Deep Learning

Sudhir Kumar Pathak1, Vishwesh Nath2, Sandip Panesar3, Kurt G. Schilling4, Juan Carlos Fernandez-Miranda3, Bennett A. Landman2,5, and Walter Schneider1
1Learning Research and Development Center, University of Pittsburgh, Pittsburgh, PA, United States, 2Electrical Engineering & Computer Science, Vanderbilt University, Nashville, TN, United States, 3Department of Neurosurgery, Stanford University, Palo Alto, CA, United States, 4Radiology, Vanderbilt University Medical Center, Nashville, TN, United States, 5Biomedical Engineering, Vanderbilt University, Nashville, TN, United States

Diffusion-weighted magnetic resonance imaging (DW-MRI) offers a unique insight on microarchitecture of the in-vivo human brain. Multiple well-known reconstruction methods that model geometrical and micro-structural properties of the tissue such as multi-tissue constrained spherical deconvolution (MT-CSD) and spherical mean technique (SMT) rely on high quality acquisitions (more than 2 shells and 45 gradient directions) which is a constraint. We propose recovery of fiber-ODFs, compartment diffusivities and volume-fractions using a two-stage deep learning framework by training on human-connectome-project dataset. The proposed approach can predict fiber-ODFs using single shell DW-MRI on a tumor patient and assess the diseased region of interest.

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

Join Here