A small number of abnormal brain connections predicts adult autism spectrum disorder

TLDR

Autism spectrum disorder is a lifelong condition with unclear neural mechanisms, and although neuroimaging classifiers have been developed, they have not been validated on independent cohorts due to over‑fitting and varying conditions. The study aims to develop a machine‑learning algorithm that identifies a small set of functional connections distinguishing ASD from typically developing individuals. The authors use a novel machine‑learning approach to select a minimal set of functional connections that discriminate ASD from TD. The classifier achieved high accuracy in a Japanese discovery cohort and generalized well to independent USA and Japanese validation cohorts, correctly distinguishing ASD from TD while failing to separate MDD and ADHD but moderately separating schizophrenia, suggesting potential for neuroimaging‑based dimensional analysis across disorders.

Abstract

Although autism spectrum disorder (ASD) is a serious lifelong condition, its underlying neural mechanism remains unclear. Recently, neuroimaging-based classifiers for ASD and typically developed (TD) individuals were developed to identify the abnormality of functional connections (FCs). Due to over-fitting and interferential effects of varying measurement conditions and demographic distributions, no classifiers have been strictly validated for independent cohorts. Here we overcome these difficulties by developing a novel machine-learning algorithm that identifies a small number of FCs that separates ASD versus TD. The classifier achieves high accuracy for a Japanese discovery cohort and demonstrates a remarkable degree of generalization for two independent validation cohorts in the USA and Japan. The developed ASD classifier does not distinguish individuals with major depressive disorder and attention-deficit hyperactivity disorder from their controls but moderately distinguishes patients with schizophrenia from their controls. The results leave open the viable possibility of exploring neuroimaging-based dimensions quantifying the multiple-disorder spectrum.