Transactions on Mass-Data Analysis of Images and Signals (ISSN:1868-6451)
Volume 9 - Number 1 - September 2018 - Page 3-17
MRI brain imagery processing software in data analysis
Alexander Bernstein, Akzhigitov Renat, Ekaterina Kondrateva, Svetlana Sushchinskaya, Irina Samotaeva, Vladislav Gaskin
1 Skolkovo Institute of Science and Technology, Moscow, Russia. 2Moscow Research and Clinical Center for Neuropsychiatry, Moscow, Russia. 3Laboratory of applied physiology of higher nervous activity, Institute of higher nervous activity and neurophysiology, RAS, Moscow, Russia.
Abstract
Modern brain scanning techniques are one of the most consistent sources of medical information which included in most diagnostic protocols. These data possess a great potential for the machine learning (ML) analysis and clinical applications of the results. MRI imaging is prevalent in brain analysis, allow-ing acquisition of structural (3D image) and functional imaging (timeseries of 3D images). Relative to the MRI data volumes these research are considered to the big data and massive analysis, connected with numerous methods of signal processing, advanced ML approaches and feature extraction. In cur-rent paper we update the surveys of most software used for the feature gen-eration, processing and analysis of MRI brain imagery. Systematizing the toolboxes, we describe common scanner characteristics, image modalities, corresponding features and their informational context, as well as common machine learning classification problems in working with the data. And as a result, we propose evaluation and comparison of conventional software tools with their detailed description.
Keywords: MRI Software Packages, MRI toolbox, Machine learning, Preprocessing, Classification, Neuroscience
Download Paper
(347 KB)