Genetic algorithms for feature selection when classifying severe chronic disorders of consciousness
| Autor: | Masayuki Murata, Frank Rattay, Kenji Leibnitz, Martin Kronbichler, Stefan Golaszewski, Betty Wutzl |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Male
Support Vector Machine Computer science Functional magnetic resonance imaging Diagnostic Radiology Machine Learning 0302 clinical medicine Functional Magnetic Resonance Imaging Medicine and Health Sciences media_common Brain Mapping 0303 health sciences Multidisciplinary medicine.diagnostic_test Radiology and Imaging Applied Mathematics Simulation and Modeling Brain Software Engineering Middle Aged Prognosis Magnetic Resonance Imaging Physical Sciences Medicine Consciousness Disorders Engineering and Technology Female Algorithms Research Article Adult Computer and Information Sciences Consciousness Imaging Techniques Science Cognitive Neuroscience media_common.quotation_subject Neuroimaging Feature selection Research and Analysis Methods 03 medical and health sciences Diagnostic Medicine Artificial Intelligence Support Vector Machines Machine learning Genetic algorithm medicine Humans Preprocessing Aged 030304 developmental biology Support vector machines Resting state fMRI business.industry Biology and Life Sciences Magnetic resonance imaging Pattern recognition Diagnostic medicine Support vector machine Cognitive Science Artificial intelligence business Mathematics 030217 neurology & neurosurgery Neuroscience |
| Zdroj: | PLoS ONE PLoS ONE, Vol 14, Iss 7, p e0219683 (2019) |
| ISSN: | 1932-6203 |
| DOI: | 10.1371/journal.pone.0219683 |
| Popis: | The diagnosis and prognosis of patients with severe chronic disorders of consciousness are still challenging issues and a high rate of misdiagnosis is evident. Hence, new tools are needed for an accurate diagnosis, which will also have an impact on the prognosis. In recent years, functional Magnetic Resonance Imaging (fMRI) has been gaining more and more importance when diagnosing this patient group. Especially resting state scans, i.e., an examination when the patient does not perform any task in particular, seems to be promising for these patient groups. After preprocessing the resting state fMRI data with a standard pipeline, we extracted the correlation matrices of 132 regions of interest. The aim was to find the regions of interest which contributed most to the distinction between the different patient groups and healthy controls. We performed feature selection using a genetic algorithm and a support vector machine. Moreover, we show by using only those regions of interest for classification that are most often selected by our algorithm, we get a much better performance of the classifier. (VLID)4337672 |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|