Autism Detection using Visual and Behavioral Data

Abstract

Diagnosing Autism Spectrum Disorder (ASD) can be difficult as there is no existing medical test for detecting Autism. The only clinical method for diagnosing ASD are standardized tests which require prolonged diagnostic time and can be expensive. Autism diagnosis can be formulated as a typical machine learning classification problem between ASD patients and a control group, which requires large datasets with different modalities to be trained on, in order to yield accurate results. However, the unavailability of such robust datasets stands as a threat to this automated diagnosis. To resolve this, we propose a method of Autism Detection using Visual and Behavioral Data. The proposed technique first relates the two datasets by generating