Meeting Banner
Abstract #1412

CNN-based classification of multiple sclerosis using BOLD venographic imaging (SWI) data

Alina Lopatina1, Renat Sibgatulin1, Stefan Ropele2, Jürgen R Reichenbach1,3, and Daniel Güllmar1
1Medical Physics Group / IDIR, Jena University Hospital, Jena, Germany, 2Department of Neurology, Medical University of Graz, Graz, Austria, 3Michael Stifel Center for Data-driven Sciences, Friedrich-Schiller-University Jena, Jena, Germany

Convolutional neural network (CNN) was proposed to identify multiple sclerosis patients and healthy subjects while susceptibility-weighted imaging (SWI) was used for MRI scans preprocessing in order to disclose important features hidden in brain venograms. Using only one two-dimensional slice for each subject makes the proposed algorithm easy-applied and useful for clinical practice.

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

Join Here