Deep learning diagnosis plus kinematic severity assessments of neurodivergent disorders
Early diagnosis of neurodivergent disorders (NDD) remains challenging. New research suggests analysing subtle movement patterns at millisecond scales holds crucial cognitive information. Using deep learning on raw kinematic data, researchers could highly accurately identify individuals with ASD, ADHD, or comorbid conditions. Additionally, new biometrics (Fano Factor, Shannon Entropy) quantifying physiological fluctuations offer quantitative subtyping by severity, providing complementary tools for early evaluation and diagnosis.
Early diagnostic assessments of neurodivergent disorders (NDD), remains a major clinical challenge. We address this problem by pursuing the hypothesis that there is important cognitive information about NDD conditions contained in the way individuals move, when viewed at millisecond time scales.
Click here to access the article.