Automated Segmentation of Multiple Sclerosis Brain Lesions at 7T
Blake Dewey 1 , Pascal Sati 1 , Snehashis Roy 2 , Luisa Vuolo 1,3 , Colin Shea 1 , Dzung Pham 2 , and Daniel S. Reich 1
National Institute of Neurological Diseases
and Stroke, National Institutes of Health, Bethesda,
Maryland, United States,
for Neuroscience and Regenerative Medicine, The Henry M.
Jackson Foundation for the Advancement of Military
Medicine, Bethesda, Maryland, United States,
of Neurology and Radiology, University of Florence,
Automatic lesion segmentation is required for analyzing
large datasets provided by multicontrast 3D
high-resolution imaging of multiple sclerosis. Although
several methods have been proposed for MRI at clinical
field strength (3T and below), 7T imaging remains
uninvestigated due to more severe image bias (B1 field
inhomogeneities, etc.) that can deeply impact the
results of the existing segmentation algorithms. In this
study, we propose an optimized multicontrast 3D
high-resolution acquisition protocol combined with the
use of advanced nonlinear bias correction and the
LesionTOADS algorithm to create robust image
segmentation and lesion load quantification at 7T.
This abstract and the presentation materials are available to members only;
a login is required.