Autism spectrum disorder (ASD) is a disorder with a wide range of symptoms that affects almost one in every 189 girls and one in every 42 boys. For a study, the researchers sought to find reliable functional brain organization markers that could differentiate between boys and females with ASD and predict symptom severity. Researchers used data from multiple neuroimaging cohorts (ASD n=773) to create a new spatiotemporal deep neural network (stDNN) that leverages spatiotemporal convolution on functional magnetic resonance imaging data to distinguish between groups. In discriminating between females and males with ASD, stDNN consistently achieved good classification accuracy. Notably, stDNN trained to discriminate between girls and males with ASD was unable to distinguish between neurotypical females and males, implying that gender differences in ASD functional brain organization differ from normative gender differences. Females and males with ASD were reliably discriminated by brain characteristics linked with motor, linguistic, and visuospatial attentional systems. Importantly, these outcomes were confirmed in a large multisite cohort and a completely separate cohort. Furthermore, brain characteristics linked to the primary motor cortex node of the motor network predicted the degree of restricted/repetitive behaviors in females with ASD but not in males. The brains of girls and boys with ASD were functionally organized differently, leading to their clinical symptoms in diverse ways, according to the reproducible outcomes. They contributed to the establishment of gender-specific ASD diagnosis and treatment options, advancing precision psychiatry.

Source:www.cambridge.org/core/journals/the-british-journal-of-psychiatry/article/deep-learning-identifies-robust-gender-differences-in-functional-brain-organization-and-their-dissociable-links-to-clinical-symptoms-in-autism/33BBC9B3ADFCC28B28081368D1CE46DC