Autism detected from 3D walking data using attention deep learning
3D Gait-Based Autism Classification Using Attention-Enhanced Deep Learning with Cross-Fold Statistical Stability Analysis
Computer Vision and Pattern RecognitionMachine Learning
Summary
Diagnosing autism early is often hard because it relies on expert judgment and time-consuming tests. This study shows that analyzing how people walk, using 3D motion data and a special kind of deep learning model called a Transformer, can identify autism with high accuracy. The authors tested their approach on two different types of walking data and showed it works better than previous methods. They also checked how stable the results are across multiple tests to ensure reliability.
What this means in practice
- •For clinical diagnostic teams: Use 3D gait analysis combined with attention-based models to aid in early ASD screening for patients in clinical settings.
- •For biomechanics software developers: Develop gait analysis tools that incorporate attention-enhanced deep learning for improving automated detection of neurological disorders.$Commercial implications: Allows creation of advanced diagnostic software products that assess autism risk from walking data more reliably.
Authors
Md Nadim Mahamood, Md Arif Shahriar, Md Parvej Sikder, Md Rasul Islam, Md Shafi Ud Doula, Md Ashraful Alam, Kamrul Hasan
Abstract
Autism Spectrum Disorder (ASD) is a neurodevelopmental condition whose early diagnosis remains challenging because conventional clinical assessments are often subjective, time-consuming, and require expert evaluation. Gait provides a promising non-invasive behavioral biomarker for auto- mated ASD screening; however, existing studies have primarily relied on single-dataset evaluations, convolutional architectures, and descriptive summaries of cross-validation performance without formally assessing fold-to-fold stability. This study addresses these gaps with an attention-enhanced Transformer framework for ASD classification, evaluated on two structurally different 3D gait feature representations: precomputed statistical gait descriptors and raw biomechanical ground-reaction- force measurements. Under five-fold cross-validation, the proposed framework achieved 99.00% accuracy, 99.02% precision, 99.00% recall, 99.00% F1-score, and 99.00% specificity on the public Kinect-based benchmark, exceeding the performance of the compared state-of-the-art methods. On the independent private force-plate dataset, it achieved mean values of 95.00% accuracy, 93.81% precision, 96.67% recall, 95.13% F1-score, and 93.33% specificity.