Weakly-supervised Autism Severity Assessment in Long Videos

TL;DR

This paper introduces a weakly-supervised video analysis method using spatio-temporal features and a shallow TCN-MLP network to detect autism and assess its severity in long, untrimmed videos, aiding clinical diagnosis.

Contribution

It presents a novel weakly-supervised approach with a specialized TCN-MLP network for autism detection and severity assessment from long videos, which is less reliant on detailed annotations.

Findings

Effective detection of autism using behavioral biomarkers.

Accurate severity categorization demonstrated on clinical videos.

Potential to assist clinicians in autism spectrum analysis.

Abstract

Autism Spectrum Disorder (ASD) is a diverse collection of neurobiological conditions marked by challenges in social communication and reciprocal interactions, as well as repetitive and stereotypical behaviors. Atypical behavior patterns in a long, untrimmed video can serve as biomarkers for children with ASD. In this paper, we propose a video-based weakly-supervised method that takes spatio-temporal features of long videos to learn typical and atypical behaviors for autism detection. On top of that, we propose a shallow TCN-MLP network, which is designed to further categorize the severity score. We evaluate our method on actual evaluation videos of children with autism collected and annotated (for severity score) by clinical professionals. Experimental results demonstrate the effectiveness of behavioral biomarkers that could help clinicians in autism spectrum analysis.