Multimodal Schizophrenia Detection by Multiclassification Analysis Aydın Ula¸ s 1⋆ , Umberto Castellani 1 , Pasquale Mirtuono 1 , Manuele Bicego 1,2 , Vittorio Murino 1,2 , Stefania Cerruti 3 , Marcella Bellani 3 , Manfredo Atzori 4 , Gianluca Rambaldelli 3 , Michele Tansella 3 , and Paolo Brambilla 4,5 1 University of Verona, Department of Computer Science, Verona, Italy 2 Istituto Italiano di Tecnologia (IIT), Genova, Italy 3 Department of Public Health and Community Medicine, Section of Psychiatry and Clinical Psychology, Inter-University Centre for Behavioural Neurosciences, University of Verona, Verona, Italy 4 IRCCS “E. Medea” Scientific Institute, Udine, Italy 5 DISM, Inter-University Centre for Behavioural Neurosciences, University of Udine, Udine, Italy Abstract. We propose a multiclassification analysis to evaluate the rel- evance of different factors in schizophrenia detection. Several Magnetic Resonance Imaging (MRI) scans of brains are acquired from two sen- sors: 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 %. Keywords: Machine learning algorithms, Magnetic resonance imaging, Support vector machines, Correlation 1 Introduction Computational neuroanatomy using magnetic resonance imaging (MRI) is a growing research field that employs image analysis methods to quantify mor- phological characteristics of different brains [5]. The ultimate goal is to identify structural brain abnormalities by comparing normal subjects (controls) with patients affected by a certain disease. Advanced computer vision and pattern ⋆ Corresponding author.