TY - JOUR
T1 - Integrating machining learning and multimodal neuroimaging to detect schizophrenia at the level of the individual
AU - Lei, Du
AU - Pinaya, Walter H. L.
AU - Young, Jonathan
AU - van Amelsvoort, Therese
AU - Marcelis, Machteld
AU - Donohoe, Gary
AU - Mothersill, David O.
AU - Corvin, Aiden
AU - Vieira, Sandra
AU - Huang, Xiaoqi
AU - Lui, Su
AU - Scarpazza, Cristina
AU - Arango, Celso
AU - Bullmore, Ed
AU - Gong, Qiyong
AU - McGuire, Philip
AU - Mechelli, Andrea
N1 - Funding Information:
European Commission, Grant/Award Number: 603196; European Research Council, Grant/Award Number: REA‐677467; National Natural Science Foundation of China, Grant/Award Numbers: 81220108013, 81501452, 81761128023, 81227002; Newton International Fellowship, Grant/Award Number: NF151455; Science Foundation Ireland, Grant/Award Number: SFI 12/1365; Wellcome Trust's Innovator Award, Grant/Award Number: 208519/Z/17/Z Funding information
Funding Information:
This study was supported by a grant from the European Commission (PSYSCAN—Translating neuroimaging findings from research into clinical practice; ID: 603196), Funds for International Cooperation and Exchange of the National Natural Science Foundation of China (grant no. 81220108013) jointly awarded to Prof. Qiyong Gong and Prof. Andrea Mechelli, National Natural Science Foundation of China (grant no. 81501452), a Wellcome Trust's Innovator Award awarded to Prof. Andrea Mechelli (208519/Z/17/Z); and a Newton International Fellowship to Dr. Du Lei (ID: NF151455). Collection of Dataset 4 was funded by Science Foundation Ireland (SFI 12/1365) and Prof. Donohoe is funded by a European Research Council Award (REA‐677467). Collection of Dataset 5 was funded by the National Natural Science Foundation of China (grant nos. 81621003,81761128023 and 81227002). The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Publisher Copyright:
© 2019 The Authors. Human Brain Mapping published by Wiley Periodicals, Inc.
PY - 2020/4/1
Y1 - 2020/4/1
N2 - Schizophrenia is a severe psychiatric disorder associated with both structural and functional brain abnormalities. In the past few years, there has been growing interest in the application of machine learning techniques to neuroimaging data for the diagnostic and prognostic assessment of this disorder. However, the vast majority of studies published so far have used either structural or functional neuroimaging data, without accounting for the multimodal nature of the disorder. Structural MRI and resting-state functional MRI data were acquired from a total of 295 patients with schizophrenia and 452 healthy controls at five research centers. We extracted features from the data including gray matter volume, white matter volume, amplitude of low-frequency fluctuation, regional homogeneity and two connectome-wide based metrics: structural covariance matrices and functional connectivity matrices. A support vector machine classifier was trained on each dataset separately to distinguish the subjects at individual level using each of the single feature as well as their combination, and 10-fold cross-validation was used to assess the performance of the model. Functional data allow higher accuracy of classification than structural data (mean 82.75% vs. 75.84%). Within each modality, the combination of images and matrices improves performance, resulting in mean accuracies of 81.63% for structural data and 87.59% for functional data. The use of all combined structural and functional measures allows the highest accuracy of classification (90.83%). We conclude that combining multimodal measures within a single model is a promising direction for developing biologically informed diagnostic tools in schizophrenia.
AB - Schizophrenia is a severe psychiatric disorder associated with both structural and functional brain abnormalities. In the past few years, there has been growing interest in the application of machine learning techniques to neuroimaging data for the diagnostic and prognostic assessment of this disorder. However, the vast majority of studies published so far have used either structural or functional neuroimaging data, without accounting for the multimodal nature of the disorder. Structural MRI and resting-state functional MRI data were acquired from a total of 295 patients with schizophrenia and 452 healthy controls at five research centers. We extracted features from the data including gray matter volume, white matter volume, amplitude of low-frequency fluctuation, regional homogeneity and two connectome-wide based metrics: structural covariance matrices and functional connectivity matrices. A support vector machine classifier was trained on each dataset separately to distinguish the subjects at individual level using each of the single feature as well as their combination, and 10-fold cross-validation was used to assess the performance of the model. Functional data allow higher accuracy of classification than structural data (mean 82.75% vs. 75.84%). Within each modality, the combination of images and matrices improves performance, resulting in mean accuracies of 81.63% for structural data and 87.59% for functional data. The use of all combined structural and functional measures allows the highest accuracy of classification (90.83%). We conclude that combining multimodal measures within a single model is a promising direction for developing biologically informed diagnostic tools in schizophrenia.
KW - functional connectivity
KW - graph theoretical analysis
KW - machine learning
KW - neuroimaging
KW - schizophrenia
KW - FUNCTIONAL CONNECTIVITY PATTERNS
KW - MATTER VOLUME ABNORMALITIES
KW - MAJOR DEPRESSIVE DISORDER
KW - SUPPORT VECTOR MACHINE
KW - SMALL-WORLD NETWORKS
KW - ULTRA-HIGH-RISK
KW - 1ST-EPISODE PSYCHOSIS
KW - BRAIN NETWORKS
KW - DISCRIMINATIVE ANALYSIS
KW - IMAGING BIOMARKERS
U2 - 10.1002/hbm.24863
DO - 10.1002/hbm.24863
M3 - Article
C2 - 31737978
SN - 1065-9471
VL - 41
SP - 1119
EP - 1135
JO - Human Brain Mapping
JF - Human Brain Mapping
IS - 5
ER -