Regional muscle load grade prediction method and device based on deep learning
By combining inertial motion capture and biomechanical simulation with deep learning, a high-fidelity labeled dataset is generated, and a deep learning model is trained to achieve real-time, low-cost assessment of muscle load in multiple regions of the whole body. This solves the problems of computational complexity and universality of traditional methods and provides interpretable prediction results.
Patent Information
- Application Number
- CN202610325169.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-03-17
- Publication Date
- 2026-04-14
- Estimated Expiration
- 2046-03-17
AI Technical Summary
Existing technologies struggle to achieve real-time, low-cost, and non-invasive assessment of muscle load across multiple regions and scenarios throughout the body. Furthermore, traditional methods are computationally complex, time-consuming, and lack versatility and interpretability.
Motion data is collected by an inertial motion capture system, and after standardized preprocessing, it is input into an individualized biomechanical simulation process to generate a high-fidelity labeled dataset. A deep learning model is then trained to achieve a mapping from joint motion to muscle load levels.
We have developed an efficient, universal, and interpretable end-to-end muscle load assessment system that solves the problem of scarce training data labels and enables fast and accurate prediction of muscle load levels.