Classification of First-Episode Schizophrenia Patients and Healthy Subjects by Automated MRI Measures of Regional Brain Volume and Cortical Thickness

The University of Melbourne, Australia
PLoS ONE (Impact Factor: 3.23). 06/2011; 6(6):e21047. DOI: 10.1371/journal.pone.0021047
Source: PubMed


Although structural magnetic resonance imaging (MRI) studies have repeatedly demonstrated regional brain structural abnormalities in patients with schizophrenia, relatively few MRI-based studies have attempted to distinguish between patients with first-episode schizophrenia and healthy controls.
Three-dimensional MR images were acquired from 52 (29 males, 23 females) first-episode schizophrenia patients and 40 (22 males, 18 females) healthy subjects. Multiple brain measures (regional brain volume and cortical thickness) were calculated by a fully automated procedure and were used for group comparison and classification by linear discriminant function analysis.
Schizophrenia patients showed gray matter volume reductions and cortical thinning in various brain regions predominantly in prefrontal and temporal cortices compared with controls. The classifiers obtained from 66 subjects of the first group successfully assigned 26 subjects of the second group with accuracy above 80%.
Our results showed that combinations of automated brain measures successfully differentiated first-episode schizophrenia patients from healthy controls. Such neuroimaging approaches may provide objective biological information adjunct to clinical diagnosis of early schizophrenia.

Download full-text


Available from: Tsutomu Takahashi,
  • Source
    • "For example, Pohl and Sabuncu (2009) obtained an accuracy of 90% in the distinction between controls and subjects presenting with a first episode of schizophrenia , employing as features the parameters of the transformation used to align segmented gray matter structures to an atlas. Abnormalities in the shape of gray matter, especially described by parameters obtained from registration or alignment processes, have been demonstrated also by other groups, utilizing thickness and volume of selected brain areas (Takayanagi et al., 2011; Borgwardt et al., 2013; Zanetti et al., 2013) or other gray matter characteristics, as gray matter density in the work of Sun et al. (2009) reaching accuracies of around 85%. Patients suffering from long lasting psychosis have also been efficiently distinguished from healthy volunteers using structural imaging and SVM. Earlier works of our group (Castellani et al., 2011; Castellani et al., 2012; Ulas et al., 2011) have showed that patients with schizophrenia were consistently classified, based on region of interest shape descriptors, with accuracies of up to over 80%. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Hemodynamic changes in the brain have been reported in major psychosis in respect to healthy controls, and could unveil the basis of structural brain modifications happening in patients. The study of first episode psychosis is of particular interest because the confounding role of chronicity and medication can be excluded. The aim of this work is to automatically discriminate first episode psychosis patients and normal controls on the basis of brain perfusion employing a support vector machine (SVM) classifier. 35 normal controls and 35 first episode psychosis underwent dynamic susceptibility contrast magnetic resonance imaging, and cerebral blood flow and volume, along with mean transit time were obtained. We investigated their behavior in the whole brain and in selected regions of interest, in particular the left and right frontal, parietal, temporal and occipital lobes, insula, caudate and cerebellum. The distribution of values of perfusion indexes were used as features in a support vector machine classifier. Mean values of blood flow and volume were slightly lower in patients, and the difference reached statistical significance in the right caudate, left and right frontal lobes, and in left cerebellum. Linear SVM reached an accuracy of 83% in the classification of patients and normal controls, with the highest accuracy associated with the right frontal lobe and left parietal lobe. In conclusion, we found evidence that brain perfusion could be used as a potential marker to classify patients with psychosis, who show reduced blood flow and volume in respect to normal controls. Copyright © 2015 Elsevier B.V. All rights reserved.
    Schizophrenia Research 04/2015; 165(1). DOI:10.1016/j.schres.2015.03.017 · 3.92 Impact Factor
  • Source
    • "Our findings of cortical thinning involving frontal, temporal and parietal cortices are consistent with cortical thickness changes which have been previously reported in schizophrenia at different phases of illness including first episode cases [22,23] as well as in those with chronic illness [24]. Crespo-Facorro et al. [22] reported significant total cortical thinning and especially involving frontal, temporal and parietal cortices in their study of 142 patients with first episode schizophrenia and compared with 83 healthy controls. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Although the genome wide supported psychosis susceptibility neurogranin (NRGN) gene is expressed in human brains, it is unclear how it impacts brain morphology in schizophrenia. We investigated the influence of NRGN rs12807809 on cortical thickness, subcortical volumes and shapes in patients with schizophrenia. One hundred and fifty six subjects (91 patients with schizophrenia and 65 healthy controls) underwent structural MRI scans and their blood samples were genotyped. A brain mapping algorithm, large deformation diffeomorphic metric mapping, was used to perform group analysis of subcortical shapes and cortical thickness. Patients with risk TT genotype were associated with widespread cortical thinning involving frontal, parietal and temporal cortices compared with controls with TT genotype. No volumetric difference in subcortical structures (hippocampus, thalamus, amygdala, basal ganglia) was observed between risk TT genotype in patients and controls. However, patients with risk TT genotype were associated with thalamic shape abnormalities involving regions related to pulvinar and medial dorsal nuclei. Our results revealed the influence of the NRGN gene on thalamocortical morphology in schizophrenia involving widespread cortical thinning and thalamic shape abnormalities. These findings help to clarify underlying NRGN mediated pathophysiological mechanisms involving cortical-subcortical brain networks in schizophrenia.
    PLoS ONE 12/2013; 8(12):e85603. DOI:10.1371/journal.pone.0085603 · 3.23 Impact Factor
  • Source
    • "c o m / l o c a t e / s c h r e s At present, only two studies have examined both GMV and CT in the early phases of psychosis (Voets et al., 2008; Takayanagi et al., 2011); these have reported discrepant findings as to whether GMV and CT alterations are expressed in similar or different areas. More specifically, Takayanagi et al. (2011) observed reduced CT and GMV in prefrontal and temporal regions in patients with a first episode of psychosis (Takayanagi et al., 2011). In contrast, Voets and colleagues showed widespread but focally distinct patterns of CT and GMV alterations in adolescents with early-onset schizophrenia (Voets et al., 2008). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Previous studies have reported alterations in grey matter volume and cortical thickness in individuals at high risk of developing psychosis and patients in the early stages of the disorder. Because these studies have typically focused on either grey matter volume or cortical thickness separately, the relationship between these two types of alterations is currently unclear. In the present investigation we used both voxel-based cortical thickness (VBCT) and voxel-based morphometry (VBM) to examine neuroanatomical differences in 21 individuals with an At Risk Mental State (ARMS) for psychosis, 26 patients with a First Episode of Psychosis (FEP) and 24 healthy controls. Statistical inferences were made at P<0.05 after correction for multiple comparisons. Cortical thinning in the right superior temporal gyrus was observed in both individuals at high risk of developing psychosis and patients with a first episode of the disorder, and therefore is likely to represent a marker of vulnerability. In contrast, the right posterior cingulate cortex showed cortical thinning in FEP patients relative to individuals at high risk, and therefore appears to be implicated in the onset of the disease. These neuroanatomical differences were expressed in terms of cortical thickness but not in terms of grey matter volume, and therefore may reflect specific cortical atrophy as opposed to variations in sulcal and gyral morphology.
    Schizophrenia Research 09/2013; 150(2-3). DOI:10.1016/j.schres.2013.08.030 · 3.92 Impact Factor
Show more