Development, Validation and Feasibility of a Yoga-Based Motor Assessment Tool (Y-MAT) for Children with ADHD: A Cross-Sectional Pilot Study

<ns3:p>Background Motor dysregulation occurs in 30–50% of children with Attention-Deficit/Hyperactivity Disorder (ADHD), reflecting impairments in neuromuscular control linked to fronto-striato-cerebellar circuitry. Conventional motor batteries may not adequately capture ADHD-specific regulatory deficits. This study aimed to develop, validate, and evaluate the feasibility of a Yoga-Based Motor Assessment Tool (Y-MAT) for children with ADHD. Methods A cross-sectional pilot study was conducted at a tertiary psychiatric centre in India. Seven motor domains implicated in ADHD were identified through a literature review and mapped to yoga-based tasks by subject experts. Content validity was assessed using Lawshe’s Content Validity Ratio (CVR). Using purposive sampling, 24 children aged 8–15 years (12 ADHD, 12 typically developing controls) underwent Y-MAT assessment. ADHD symptom severity and adaptive functioning were evaluated using the ADHD Rating Scale (ADHD-RS) and Vineland Social Maturity Scale. Video recordings were independently scored by two blinded raters. An AI prototype employing pose estimation (Google MediaPipe and YOLOv8) was piloted on selected domains. Results Five yoga-based tasks achieved CVR = 1.00 and constituted the final Y-MAT. Feasibility metrics demonstrated high acceptability, demand, implementation fidelity, and practicality. Children with ADHD scored significantly lower than controls (human rating: 10.8 ± 7.1 vs 23.8 ± 4.0, p < 0.001; AI scoring: 12.5 ± 5.9 vs 19.1 ± 2.2, p = 0.002). Inter-rater reliability ranged from Kendall’s W = 0.23 to 0.55. Y-MAT total scores showed significant negative correlations with ADHD symptom severity. ROC analysis demonstrated excellent discriminatory validity (AUC = 0.910 human scoring; AUC = 0.896 AI scoring). Conclusions Y-MAT demonstrates preliminary validity and strong feasibility as a culturally grounded motor regulation assessment for children with ADHD. The AI prototype provides proof of concept for objective, scalable screening. Larger multi-centre studies are needed to establish psychometric robustness and normative thresholds.</ns3:p>