Investigating brain structural patterns in first episode psychosis and schizophrenia using MRI and a machine learning approach

Author(s)

de Moura, Adriana Miyazaki ^[1] ; Lopez Pinaya, Walter Hugo ^[1] ; Gadelha, Ary ^[2] ; Zugman, Andre ^[2] ; Noto, Cristiano ^[2] ; Cordeiro, Quirino ^{[2, 3]} ; Belangero, Sintia Iole ^{[4, 2]} ; Jackowski, Andrea P. ^[2] ; Bressan, Rodrigo A. ^[2] ; Sato, Joao Ricardo ^{[2, 1]}

Affiliation

^[1] Fed Univ ABC, Ctr Math Computat & Cognit, Rua Arcturus, 03 Jardim Antares, BR-09606070 Sao Bernardo Do Campo, SP - Brazil

^[2] Univ Fed Sao Paulo, Dept Psychiat, Sao Paulo - Brazil

^[3] Med Sch Santa Casa Sao Paulo, Dept Psychiat, Sao Paulo - Brazil

^[4] Univ Fed Sao Paulo, Dept Morphol & Genet, Sao Paulo - Brazil

• Document type: Journal article
• Source: PSYCHIATRY RESEARCH-NEUROIMAGING; v. 275, p. 14-20, MAY 30 2018.
• Web of Science Citations: 0

Abstract

In this study, we employed the Maximum Uncertainty Linear Discriminant Analysis (MLDA) to investigate whether the structural brain patterns in first episode psychosis (FEP) patients would be more similar to patients with chronic schizophrenia (SCZ) or healthy controls (HC), from a schizophrenia model perspective. Brain regions volumetric data were estimated by using MRI images of SCZ and FEP patients and HC. First, we evaluated the MLDA performance in discriminating SCZ from controls, which provided a score based on a model for changes in brain structure in SCZ. In the following, we compared the volumetric patterns of FEP patients with patterns of SCZ and healthy controls using these scores. The FEP group had a score distribution more similar to patients with schizophrenia (p-value = .461; Cohen's d = -.15) in comparison with healthy subjects (p-value = .003; Cohen's d = .62). Structures related to the limbic system and the circuitry involved in goal-directed behaviours were the most discriminant regions. There is a distinct pattern of volumetric changes in patients with schizophrenia in contrast to healthy controls, and this pattern seem to be detectable already in FEP. (AU)

• FAPESP's process: 13/10498-6 - Machine learning in neuroimaging: development of methods and clinical applications in psychiatric disorders
• Grantee: João Ricardo Sato
• Support Opportunities: Regular Research Grants
• FAPESP's process: 13/08531-5 - High risk cohort study for psychiatric disorders in childhood: 3-year follow-up neuroimaging study
• Grantee: Andrea Parolin Jackowski
• Support Opportunities: Regular Research Grants
• FAPESP's process: 14/07280-1 - Search for genetic markers of risk, progression and response to treatment
• Grantee: Síntia Iole Nogueira Belangero
• Support Opportunities: Regular Research Grants
• FAPESP's process: 13/05168-7 - Development of Deep Belief Networks to support the diagnosis of psychiatric disorders
• Grantee: Walter Hugo Lopez Pinaya
• Support Opportunities: Scholarships in Brazil - Doctorate