Physics Augmented Tuple Transformer for Autism Severity-Level Detection

Early diagnosis of autism spectrum disorder (ASD) is an effective and favorable step toward enhancing the health and well-being of children with ASD. Manual ASD diagnosis testing is labor-intensive, complex, and prone to human error due to several factors contaminating the results. This article proposes a novel framework that exploits the laws of physics for ASD severity recognition. The proposed physics-informed neural network architecture encodes the behavior of the subject extracted by observing a part of the skeleton-based motion trajectory in a higher dimensional latent space. Two decoders, namely, physics-based and nonphysics-based decoder, use this latent embedding and predict the future motion patterns. The physics branch leverages the laws of physics that apply to a skeleton sequence in the prediction process while the nonphysics-based branch is optimized to minimize the difference between the predicted and actual motion of the subject. A classifier also leverages the same latent space embeddings to recognize the ASD severity. This dual generative objective explicitly forces the network to compare the actual behaviors of the subject with the general normal behaviors of children that are governed by the laws of physics, aiding the ASD recognition task. The proposed method attains state-of-the-art performance on multiple ASD diagnosis benchmarks. To illustrate the utility of the proposed framework beyond the task ASD diagnosis, we conduct a third experiment using a publicly available benchmark for the task of fall prediction and demonstrate the superiority of our model.