Answers are grounded in this paper and cite their sources.
Assesses structural integrity of AI-augmented orthotic devices, suggesting improved safety and performance in children.
Additive Manufacturing (AM) enables the development of lightweight, high-performance, and personalised paediatric sports support systems, but ensuring structural integrity under dynamic and high-impact loading remains challenging. This study proposes an AI-Augmented Structural Integrity Evaluation (AI-ASIE) framework that integrates Deep Neural Networks (DNNs), Finite Element Analysis (FEA), and SHapley Additive exPlanations (SHAP) to predict and interpret failure risks in AM orthotic devices. The DNN is trained using experimental and simulated mechanical response data from multiple orthotic designs and materials, while FEA provides physics-based validation. The hybrid framework identifies structural weaknesses associated with stress concentration, fatigue, and load transfer, and SHAP analysis explains the influence of loading conditions, geometry, and material properties on device performance. Compared with FEA alone, AI-ASIE achieved 92.4% prediction accuracy (95% CI: 89.8–95.0%), a mean absolute error of 4.6%, RMSE of 5.2%, and reduced false-negative failure predictions by 36.0%. The framework also enabled 20–24% weight reduction while maintaining stress levels of 10–35 MPa, fatigue probabilities of 0.5–45%, and energy absorption of 7–38 J. These results demonstrate the potential of AI-augmented structural assessment to improve the safety, durability, and performance of paediatric orthotic devices.
No takes yet. Share an insight, caveat, or question.
Yin et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: