Abstract
The aim of this study is to establish a reliable and widely applicable muscle strength (MS) estimation model based on the Mechanomyography (MMG). Seven healthy male volunteers were recruited to collect MMG and MS during the isometric contraction of their triceps. For MMG, 18 features were extracted. For the extreme gradient boosting (XGBoost) model and the quadratic polynomial (QP) model, the feature combination with the best estimation result was selected. The MS estimation performance of the XGBoost model and the QP model were compared. The performance of the QP model on the estimation of MS in different frequencies, different fatigue states and time periods was evaluated by using t-test. The results showed that when the number of features exceeds three, the model estimation accuracy has not improved significantly; and there was no significant difference in the estimation result of MS between the two models (p < 0.05), though the QP model was slightly better. The normalized root mean square error (NRMSE) and goodness of fit R of the MS estimation by the QP model were: 0.1343 ± 0.0296 and 0.8273 ± 0.0376. There was no significant difference in the MS estimation results in different conditions (p < 0.05).
Highlights
Muscle strength (MS) estimation has been widely used in many fields, such as exoskeleton [1], prosthetic control [2], rehabilitation robot [3], and muscle disease research [4]
The results showed that when the number of features exceeds three, the model estimation accuracy has not improved significantly; and there was no significant difference in the estimation result of muscle strength (MS) between the two models (p < 0.05), though the quadratic polynomial (QP) model was slightly better
In this study, features of MMG were extracted and selected for the XGBoost model and the QP model, and a combination of RMS, waveform length (WL), and difference absolute standard deviation value (DASD) was used to compare the performance of XGBoost model and QP model on MS estimation
Summary
Muscle strength (MS) estimation has been widely used in many fields, such as exoskeleton [1], prosthetic control [2], rehabilitation robot [3], and muscle disease research [4]. MMG is a low frequency signal generated by the lateral vibration of muscle fibers during muscle contraction, which can reflect the mechanical characteristics of muscle contraction. Compared with EMG, MMG has certain advantages: a) due to the propagation characteristics of MMG in muscle tissue, the MMG sensors do not need to be accurately placed [5, 6]; b) MMG is a mechanical signal which will not be influenced by the change of the skin impedance due to sweating [7]. As a counterpart of EMG, MMG is widely used in the researches of the action pattern recognition [7,8,9], muscle fatigue [10,11,12] and so on. Lei et al [15] selected RMS and spectral variance as MMG features, and find that
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have