Knee Muscle Force Estimating Model Using Machine Learning Approach

Anurag Sohane,Ravinder Agarwal

DOI: https://doi.org/10.1093/comjnl/bxaa160

2020-12-14

The Computer Journal

Abstract:Abstract Various simulation type tools and conventional algorithms are being used to determine knee muscle forces of human during dynamic movement. These all may be good for clinical uses, but have some drawbacks, such as higher computational times, muscle redundancy and less cost-effective solution. Recently, there has been an interest to develop supervised learning-based prediction model for the computationally demanding process. The present research work is used to develop a cost-effective and efficient machine learning (ML) based models to predict knee muscle force for clinical interventions for the given input parameter like height, mass and angle. A dataset of 500 human musculoskeletal, have been trained and tested using four different ML models to predict knee muscle force. This dataset has obtained from anybody modeling software using AnyPyTools, where human musculoskeletal has been utilized to perform squatting movement during inverse dynamic analysis. The result based on the datasets predicts that the random forest ML model outperforms than the other selected models: neural network, generalized linear model, decision tree in terms of mean square error (MSE), coefficient of determination (R2), and Correlation (r). The MSE of predicted vs actual muscle forces obtained from the random forest model for Biceps Femoris, Rectus Femoris, Vastus Medialis, Vastus Lateralis are 19.92, 9.06, 5.97, 5.46, Correlation are 0.94, 0.92, 0.92, 0.94 and R2 are 0.88, 0.84, 0.84 and 0.89 for the test dataset, respectively.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the long calculation time, muscle redundancy and low cost - effectiveness of existing methods in estimating human knee joint muscle strength during dynamic movement. The author proposes a machine - learning - based method to predict knee joint muscle strength, aiming to improve the prediction efficiency and cost - effectiveness. Specifically, the study uses four different machine - learning models (neural network, random forest, generalized linear model and decision tree), and trains and tests a data set containing input parameters such as height, weight and angle to predict knee joint muscle strength. The study found that the random forest model performs best in terms of prediction accuracy, and its mean square error (MSE), coefficient of determination (\(R^2\)) and correlation (\(r\)) are as follows: - Biceps Femoris (BF): MSE = 19.92, \(r\) = 0.94, \(R^2\) = 0.88 - Rectus Femoris (RF): MSE = 9.06, \(r\) = 0.92, \(R^2\) = 0.84 - Vastus Medialis (VM): MSE = 5.97, \(r\) = 0.92, \(R^2\) = 0.84 - Vastus Lateralis (VL): MSE = 5.46, \(r\) = 0.94, \(R^2\) = 0.89 These results indicate that the random forest model has high accuracy and reliability in predicting knee joint muscle strength and can be used as an effective tool for clinical intervention. In addition, this model also shows significant advantages in processing speed, and it only takes 2 - 3 seconds to complete the prediction, which is much lower than the 120 - 160 seconds required by traditional methods. Therefore, this study provides an efficient and economical solution, which is helpful to improve the prediction and evaluation of knee joint muscle strength.

Knee Muscle Force Estimating Model Using Machine Learning Approach

Predicting Knee Joint Contact Forces During Normal Walking Using Kinematic Inputs with A Long-Short Term Neural Network

Muscle force estimation from lower limb EMG signals using novel optimised machine learning techniques

Prediction of Knee Joint Moment by Surface Electromyography of the Antagonistic and Agonistic Muscle Pairs

On the prediction of tibiofemoral contact forces for healthy individuals and osteoarthritis patients during gait: a comparative study of regression methods

A Machine Learning Approach for Predicting Pedaling Force Profile in Cycling

Prediction of medial knee contact force using multisource fusion recurrent neural network and transfer learning

Estimation Of Knee Extension Force Using Mechanomyography Signals Detected Through Clothing

SEMG-based multifeatures and predictive model for knee-joint-angle estimation

Prediction of Knee Joint Compartmental Loading Maxima Utilizing Simple Subject Characteristics and Neural Networks

Predicting Knee Joint Contact Force Peaks During Gait Using a Video Camera or Wearable Sensors

An Automated and Robust Tool for Musculoskeletal and Finite Element Modeling of the Knee Joint

The bionic neural network for external simulation of human locomotor system

Application of Machine Learning Methods to Investigate Joint Load in Agility on the Football Field: Creating the Model, Part I

Implementing Machine Learning approaches for accelerated prediction of bone strain in acetabulum of a hip joint

Physics-informed Deep Learning for Musculoskeletal Modelling: Predicting Muscle Forces and Joint Kinematics from Surface EMG

A Subject-Specific Neuromusculoskeletal Model for Muscle Force Estimation Used in Rehabilitation

Double Condyles Contact Force Estimation of Knee Joint Based on Hill-Type Model

Predicting the Performance and Adaptation of Artificial Elbow Due to Effective Forces using Deep Learning

Physics-Informed Deep Learning for Musculoskeletal Modeling: Predicting Muscle Forces and Joint Kinematics from Surface EMG

Time series generative adversarial network for muscle force prognostication using statistical outlier detection