Weakly-supervised Autism Severity Assessment in Long Videos
This addresses autism diagnosis and monitoring for clinicians by providing a video-based assessment tool, though it appears incremental as it builds on existing biomarker approaches.
The researchers tackled autism severity assessment by developing a weakly-supervised method using spatio-temporal features from long videos to detect typical and atypical behaviors, and a TCN-MLP network to categorize severity scores, with evaluation on clinically annotated videos showing effectiveness for clinical analysis.
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.