Physics-Informed Deep Learning for Muscle Force Prediction With Unlabeled sEMG Signals
Open Access
- 14 March 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Systems and Rehabilitation Engineering
- Vol. 32 (15344320) , 1246-1256
Abstract
Computational biomechanical analysis plays a pivotal role in understanding and improving human movements and physical functions. Although physics-based modeling methods can interpret the dynamic interaction between the neural drive to muscle dynamics and joint kinematics, they suffer from high computational latency. In recent years, data-driven methods have emerged as a promising alternative due to their fast execution speed, but label information is still required during training, which is not easy to acquire in practice. To tackle these issues, this paper presents a novel physics-informed deep learning method to predict muscle forces without any label information during model training. In addition, the proposed method could also identify personalized muscle-tendon parameters. To achieve this, the Hill muscle model-based forward dynamics is embedded into the deep neural network as the additional loss to further regulate the behavior of the deep neural network. Experimental validations on the wrist joint from six healthy subjects are performed, and a fully connected neural network (FNN) is selected to implement the proposed method. The predicted results of muscle forces show comparable or even lower root mean square error (RMSE) and higher coefficient of determination compared with baseline methods, which have to use the labeled surface electromyography (sEMG) signals, and it can also identify muscle-tendon parameters accurately, demonstrating the effectiveness of the proposed physics-informed deep learning method.Keywords
All Related Versions
Funding Information
- United Kingdom Research and Innovation (UKRI) Horizon Europe Guarantee (EP/Y027930/1)
- Royal Society (IEC/NSF/211360)
- European Union (EU) Marie Curie Individual Fellowship (101023097)
- China Scholarship Council (202208320117)
This publication has 43 references indexed in Scilit:
- A multi-resolution physics-informed recurrent neural network: formulation and application to musculoskeletal systemsComputational Mechanics, 2023
- Thermodynamics-informed neural networks for physically realistic mixed realityComputer Methods in Applied Mechanics and Engineering, 2023
- A novel sEMG-based force estimation method using deep-learning algorithmComplex & Intelligent Systems, 2021
- An energy approach to the solution of partial differential equations in computational mechanics via machine learning: Concepts, implementation and applicationsComputer Methods in Applied Mechanics and Engineering, 2020
- Deep Learning for Musculoskeletal Force PredictionAnnals of Biomedical Engineering, 2018
- The Deep Ritz Method: A Deep Learning-Based Numerical Algorithm for Solving Variational ProblemsCommunications in Mathematics and Statistics, 2018
- Optimization-based dynamic prediction of kinematic and kinetic patterns for a human vertical jump from a squatting positionMultibody System Dynamics, 2015
- Modeling and computational issues in the inverse dynamics simulation of triple jumpMultibody System Dynamics, 2013
- Development of a comprehensive musculoskeletal model of the shoulder and elbowMedical & Biological Engineering & Computing, 2011
- Evaluation of predicted knee‐joint muscle forces during gait using an instrumented knee implantJournal of Orthopaedic Research, 2009