Multimodal Schizophrenia Detection by Multiclassification Analysis
We propose a multiclassification analysis to evaluate the relevance of different factors in schizophrenia detection. Several Magnetic Resonance Imaging (MRI) scans of brains are acquired from two sensors: morphological and diffusion MRI. Moreover, 14 Region Of Interests (ROIs) are available to focus the analysis on specific brain subparts. All information is combined to train three types of classifiers to distinguish between healthy and unhealthy subjects. Our contribution is threefold: (i) the classification accuracy improves when multiple factors are taken into account; (ii) proposed procedure allows the selection of a reduced subset of ROIs, and highlights the synergy between the two modalities; (iii) correlation analysis is performed for every ROI and modality to measure the information overlap using the correlation coefficient in the context of schizophrenia classification. We see that we achieve 85.96 % accuracy when we combine classifiers from both modalities, whereas the highest performance of a single modality is 78.95 %.
KeywordsMachine learning algorithms Magnetic resonance imaging Support vector machines Correlation
- 3.Cheng, D.S., Bicego, M., Castellani, U., Cerruti, S., Bellani, M., Rambaldelli, G., Atzori, M., Brambilla, P., Murino, V.: Schizophrenia classification using regions of interest in brain mri. In: Proceedings of Intelligent Data Analysis in Biomedicine and Pharmacology, IDAMAP 2009, pp. 47–52 (2009)Google Scholar
- 6.Kuncheva, L.I.: Combining pattern classifiers: methods and algorithms. Wiley-Interscience (2004)Google Scholar
- 8.Petrakos, M., Kannelopoulos, I., Benediktsson, J.A., Pesaresi, M.: The effect of correlation on the accuracy of the combined classifier in decision level fusion. In: Proceedings of the IEEE International Geo-science and Remote Sensing Symposium, IGARSS 2000, vol. 6, pp. 2623–2625 (2000)Google Scholar
- 9.Rujescu, D., Collier, D.A.: Dissecting the many genetic faces of schizophrenia. Epidemiologia e Psichiatria Sociale 18(2), 91–95 (2009)Google Scholar
- 13.Ulaş, A., Yıldız, O.T., Alpaydın, E.: Eigenclassifiers for combining correlated classifiers. Information Sciences (accepted, 2011)Google Scholar