Multivariate classification of drug-naive obsessive-compulsive disorder patients and healthy controls by applying an SVM to resting-state functional MRI data

• Published in: BMC psychiatry 2019-07, Vol.19 (1), p.210-210, Article 210
• Main Authors: Yang, Xi, Hu, Xinyu, Tang, Wanjie, Li, Bin, Yang, Yanchun, Gong, Qiyong, Huang, Xiaoqi
• Format: Article
• Language: English
• Subjects: Adult; Brain - physiopathology; Brain Mapping - methods; Care and treatment; Case-Control Studies; Classification; Diagnosis; Drug addiction; Drug-naive; Female; Fractional amplitude of low-frequency fluctuation; Functional magnetic resonance imaging; Humans; Learning algorithms; Limbic System - diagnostic imaging; Limbic System - physiopathology; Machine learning; Magnetic resonance imaging; Magnetic Resonance Imaging - methods; Magnetic Resonance Imaging - statistics & numerical data; Male; Medical imaging; Mental disorders; Multivariate Analysis; Multivariate classification; Neuroses; Obsessive compulsive disorder; Obsessive-Compulsive Disorder - diagnostic imaging; Obsessive-Compulsive Disorder - pathology; Parietal lobe; Parietal Lobe - diagnostic imaging; Parietal Lobe - physiopathology; Patients; Pattern recognition; Psychiatrists; Psychiatry; Psychotherapy; Rest - psychology; Resting-state fMRI; Risk factors; Software; Superior parietal lobule; Superior temporal gyrus; Support vector machine; Support Vector Machine - statistics & numerical data; Support vector machines; Temporal gyrus; Temporal Lobe - diagnostic imaging; Temporal Lobe - physiopathology; Young Adult
• ISSN: 1471-244X; 1471-244X
• DOI: 10.1186/s12888-019-2184-6

Previous resting-state functional magnetic resonance imaging (rs-fMRI) studies have revealed intrinsic regional activity alterations in obsessive-compulsive disorder (OCD), but those results were based on group analyses, which limits their applicability to clinical diagnosis and treatment at the level of the individual. We examined fractional amplitude low-frequency fluctuation (fALFF) and applied support vector machine (SVM) to discriminate OCD patients from healthy controls on the basis of rs-fMRI data. Values of fALFF, calculated from 68 drug-naive OCD patients and 68 demographically matched healthy controls, served as input features for the classification procedure. The classifier achieved 72% accuracy (p ≤ 0.001). This discrimination was based on regions that included the left superior temporal gyrus, the right middle temporal gyrus, the left supramarginal gyrus and the superior parietal lobule. These results indicate that OCD-related abnormalities in temporal and parietal lobe activation have predictive power for group membership; furthermore, the findings suggest that machine learning techniques can be used to aid in the identification of individuals with OCD in clinical diagnosis.