Implementing a surface EMG model with accurate parameters and a force output

Abstract

A surface electromyogram model has been implemented with both electrical and mechanical outputs. The model has been populated with parameter values based on random distributions about experimentally observed averages. Motor unit firing rates and recruitment threshold have been defined independently for type I and type II muscle fibres. This has led to non-linear inputs to the sEMG model. The observed model outputs have been compared with experimental data. Both the electrical and mechanical outputs are shown to increase linearly with force. The rates of increase are comparable between the simulated and experimental data sets. This model is generating life-like sEMG signals and may be used to study the relationship between muscle physiology and the sEMG signal.

Implementation and experimental validation of surface electromyogram and force model of Tibialis Anterior muscle for examining muscular factors

“…While more natural representations of volume conductor have been reported, 13,19,21 in this study we have investigated the model only for isometric contractions due to the limits in the assumptions. 12,25,27,29 Another significant inference is that the shortening of muscle fibres can impact the Median Frequency (MDF) 27 and could explain the reduced variability in the simulated compared with experimental sEMG. It has been shown that pennate muscles can still undergo a decrease in fibre length and increased pennation angle under isometric contraction.…”

“…The model described in this study is an adaptation from the sEMG model of biceps brachii. 4 It incorporates the following from the earlier model: 4 statistical distribution of parameters, 25 difference between fibre type, non-linear MU recruitment [25][26][27] and various sizes of MUs. 27,28 The significance of this model is that it has added the following three new features: twitch force, pennation angle and single fibre action potential (SFAP).…”

Generation of synthetic surface electromyography signals under fatigue conditions for varying force inputs using feedback control algorithm

“…Both mechanical and electrical knowledge is required to implement the model. 27,33 The model is based on parameter values, which are random distributions about experimentally proven averages. The model provides details about the current and voltage produced at each MU during muscular activities.…”

“…K i = 2.5 × 10 −4 and K p = 2 × 10 −6 . 33 The coefficient of variation (CV) for force over time, K d is the variable used to incorporate the increase in force during the contraction, which influences the ability of the algorithm to make the necessary adjustments. The algorithm adjusts itself to achieve the target force by varying the REI.…”

“…For a muscle fiber, the difference between the extra- and intracellular current distributions can be modeled as the second-order spatial derivative of AP. The current distribution 33,34 of the fiber is given as…”