Developing a Behavioral Phenotype Screening to Analyze Sex-Differential Risk in Autism

Overview

Autism Spectrum Disorder is the fastest growing neurodevelopmental disorder with four times the prevalence in males than females. Yet, there has been little focus on the differences in the manifestation and severity of autism symptoms across genders. Our study designed an objective metric to quantify behavioral differences in girls and boys diagnosed with autism. The current gold standard, Autism Diagnostic Observation Schedule (ADOS), suffers from the subjectiveness of clinician evaluations, inability to distinguish symptom severities, and failure to detect autism early because it only evaluates existing conditions of children. We evaluated 70+ autism symptoms not currently assessed in the ADOS through computer vision-based motion technology, automated speech analysis, and facial expression recognition to analyze sexual dimorphism in autism. We hypothesized that boys with autism would exhibit more abnormal facial expression than girls, while girls would maintain statistically significant higher distance to adults than boys and have more deficits in vocalization. Using SPSS, we conducted t-tests with gender as the grouping variable to see how boys and girls differ in traits affected in autism. We identified 10 quantitative indices of ASD symptoms where boys and girls differed in traits. The strongest gender-based divergence is in eye contact where girls make significantly higher eye contact. Quantifying traits like this can help improve diagnostics and ensure they reflect the reality of how girls and boys differentially experience autism. Overall, this study provides a foundation for automating autism diagnosis through digital phenotyping and identifies patterns of sexual dimorphism in autism.

Competition history

• JSHS 2022 Category not listed

Resources

• National JSHS abstract book ↗

Related projects

Closest projects by meaning, across every fair and year in the corpus.