Early-stage autism diagnosis using action videos and contrastive feature learning.

Autism, also known as Autism Spectrum Disorder (or ASD), is a neurological disorder. Its main symptoms include difficulty in verbal/non-verbal communication and rigid/repetitive behavior. These symptoms are often indistinguishable from a normal (control) individual due to which this disorder remains undiagnosed in early childhood, thus leading to a delayed treatment. Since the learning curve is steep during the initial years, an early diagnosis of autism would allow to make an early intervention, which in turn would positively affect the growth of an autistic child. Further, the traditional methods of autism diagnosis require multiple visits to a specialized doctor; however, this process is generally time-consuming. In this paper, we present a learning-based approach to automate autism diagnosis using simple and small action video clips of subjects. This task is particularly challenging because the amount of annotated data available is small, and the variations among samples from the two categories (ASD and control) are generally indistinguishable. This is also evident from poor performance of a baseline binary classifier on this task. To address this, we propose to adopt contrastive feature learning for the first time on this task and demonstrate a significant increase in the prediction accuracy. We further validate this by conducting thorough experimental analyses in both self-supervised and supervised setups on two publicly available datasets. We have also released our codes and pre-trained models for reproducibility. [ABSTRACT FROM AUTHOR]