Screening of autism based on task-free fMRI using graph theoretical approach

Studies on autism spectrum disorder (ASD) have indicated several dysfunctions in the structure, and functional organization of the brain. However, findings have not been established as a general diagnostic tool yet. In this regard, current study proposed an automatic screening method for recognition of ASDs from healthy controls (HCs) based on their brain functional abnormalities. In this paradigm, brain functional networks of 60 adolescent and young adult males (29 ASDs and 31 HCs) were estimated from subjects' task-free fMRI data. Then, autism screening was developed based on characteristics of the functional networks using the following steps: A) local and global parameters of the brain functional network were calculated using graph theory. B) network parameters of the ASDs were statistically compared to the HCs. C) significantly altered parameters were used as input features of the screening system. D) performance of the system was verified using various classification techniques. The support vector machine showed superiority to others with an accuracy of 92%. Subsequently, reliability of the results was examined using an independent dataset including 20 ASDs and 20 HCs. Our findings suggest that local parameters of the brain functional network, despite the individual variability, can potentially be used for autism screening.