New eye tracking metrics system: the value in early diagnosis of autism spectrum disorder.

Background: Eye tracking (ET) is emerging as a promising early and objective screening method for autism spectrum disorders (ASD), but it requires more reliable metrics with enhanced sensitivity and specificity for clinical use.

Methods: This study introduces a suite of novel ET metrics: Area of Interest (AOI) Switch Counts (ASC), Favorable AOI Shifts (FAS) along self-determined pathways, and AOI Vacancy Counts (AVC), applied to toddlers and preschoolers diagnosed with ASD. The correlation between these new ET metrics and Autism Diagnostic Observation Schedule, Second Edition (ADOS-2) scores via linear regression and sensitivity and specificity of the cut-off scores were assessed to predict diagnosis.

Results: Our findings indicate significantly lower FAS and ASC and higher AVC (P<0.05) in children with ASD compared to their non-ASD counterparts within this high-risk cohort; the significance was not seen in total fixation time neither pupil size (p > 0.05). Furthermore, FAS was negatively correlated with ADOS-2 total scores and social affect (SA) subscale (p < 0.05). Among these new ET metrics, AVC yielded the best sensitivity 88-100% and specificity 80-88% with cut off score 0.305-0.306, followed by FAS and ASC to separate ASD from non-ASD for diagnosis.

Conclusions: This study confirms the utility of innovative ET metrics-FAS, AVC, and ASC-which exhibit markedly improved sensitivity and specificity, enhancing ASD screening and diagnostic processes.