Anthropometry and diagnostic aware deep learning for exercise assessment
Open Access
- 6 February 2026
- journal article
- Published by Frontiers Media SA in Frontiers in Medical Technology
- Vol. 7
Abstract
Background: Correct technique during strength exercises such as squats and Romanian deadlifts (RDLs) is fundamental for performance and injury prevention. Objective: We introduce ADA (Anthropometry and Diagnostic Aware), a multimodal deep-learning framework that integrates IMU kinematics with anthropometric and diagnostic features to classify movement quality and predict movement related risk. Methods: Seventeen-sensor IMU data were collected from 15 healthy subjects performing correct and incorrect squat and RDL trials. A CNN-LSTM branch processed kinematic sequences and a fully connected branch processed static anthropometric/diagnostic inputs; feature fusion used attention weighting. Results: Incorporating anthropometry and diagnostic context increased sequence-level accuracy from 86.5% (kinematics only) to 94.8% (ADA) and enabled binary risk prediction at 97.8%. Personalized (transfer learning) fine tuning further improved accuracies (mean gains 3%–5% depending on window length). Conclusion: ADA demonstrates that subject-specific static features improve movement quality classification and risk stratification, supporting wearable-based personalized feedback in training and rehabilitation.Keywords
This publication has 24 references indexed in Scilit:
- Improved convolutional neural network for precise exercise posture recognition and intelligent health indicator predictionScientific Reports, 2025
- Muscle Force Dynamics Across Increasing Squat Intensity Conditions in Elite PowerliftersScandinavian Journal of Medicine & Science in Sports, 2025
- Revolutionizing Sports: The Role of Wearable Technology and AI in Training and Performance AnalysisQuality in Sport, 2025
- Transfer learning-enhanced CNN-GRU-attention model for knee joint torque predictionFrontiers in Bioengineering and Biotechnology, 2025
- The Impact of Stance Width on Kinematics and Kinetics During Maximum Back SquatsJournal of Strength & Conditioning Research, 2024
- A Biomechanical Comparison Between a High and Low Barbell Placement on Net Joint Moments, Kinematics, Muscle Forces, and Muscle-Specific Moments in 3 Repetition Maximum Back SquatsJournal of Strength & Conditioning Research, 2024
- Lower limb squat biomechanics and select clinical measures in chronic ankle instabilityClinical Biomechanics, 2024
- Hybrid CNN-GRU Model for Exercise Classification Using IMU Time-Series DataJournal of Machine Intelligence and Data Science, 2024
- Comparison of markerless and marker-based motion capture systems using 95% functional limits of agreement in a linear mixed-effects modelling frameworkScientific Reports, 2023
- Squatting Kinematics and Kinetics and Their Application to Exercise PerformanceJournal of Strength & Conditioning Research, 2010