Using recurrent artificial neural network model to estimate voluntary elbow torque in dynamic situations

Abstract

Muscle modelling is an important component of body segmental motion analysis. Although many studies had focused on static conditions, the relationship between electromyographic (EMG) signals and joint torque under voluntary dynamic situations has not been well investigated. The aim of this study was to investigate the performance of a recurrent artificial neural network (RANN) under voluntary dynamic situations for torque estimation of the elbow complex. EMG signals together with kinematic data, which included angle and angular velocity, were used as the inputs to estimate the expected torque during movement. Moreover, the roles of angle and angular velocity in the accuracy of prediction were investigated, and two models were compared. One model used EMG and joint kinematic inputs and the other model used only EMG inputs without kinematic data. Six healthy subjects were recruited, and two average angular velocities (60 degrees s(-1) and 90 degrees s(-1)) with three different loads (0 kg, 1 kg, 2 kg) in the hand position were selected to train and test the RANN between 90 degrees elbow flexion and full elbow extension (0 degrees). After training, the root mean squared error (RMSE) between expected torque and predicted torque of the model, with EMG and joint kinematic inputs in the training data set and the test data set, were 0.17 +/- 0.03 Nm and 0.35 +/- 0.06 Nm, respectively. The RMSE values between expected torque and predicted torque of the model, with only EMG inputs in the training data set and the test set, were 0.57 +/- 0.07 Nm and 0.73 +/- 0.11 Nm, respectively. The results showed that EMG signals together with kinematic data gave significantly better performance in the joint torque prediction; joint angle and angular velocity provided important information in the estimation of joint torque in voluntary dynamic movement.

Estimation of elbow flexion force during isometric muscle contraction from mechanomyography and electromyography

“…Among the many variations of the back propagation training methods, back propagation with the LevenbergMarquart (LM) algorithm was selected for producing the fastest convergences for medium-sized neural networks up to hundreds of neurons [32]. Its formula can be expressed as follows:…”

Lower extremity joint torque predicted by using artificial neural network during vertical jump

“…They concluded that ANN model performed reasonably well in estimating joint torque after back-propagation learning, despite of their results not aiming at predicting joint torques but those relationships. Very similar to the study of Luh et al (1999) on elbow joint torque estimating from EMG and kinematic data, Song and Tong (2005) used recurrent artificial neural network (RANN) model to estimate voluntary elbow torque in dynamic situations. EMG signals combined together with kinematic data (angle and angular velocity) were used as the inputs to estimate the expected torque during movement.…”

Human lower extremity joint moment prediction: A wavelet neural network approach

“…Predicted muscle activities were then used by a Hill-type model in order to estimate muscle forces and elbow joint torque. (Song and Tong, 2005) also investigated a recurrent artificial neural network (RANN) for elbow torque estimation using EMG data, elbow joint angle and angular velocity. (Hahn, 2007) used a three-layer FFANN to predict isokinetic knee extensor and flexor torque based on age, gender, height, body mass, EMG signals, joint position and joint velocity.…”