Motoneuron-driven computational muscle modelling with motor unit resolution and subject-specific musculoskeletal anatomy

Abstract

The computational simulation of human voluntary muscle contraction is possible with EMG-driven Hill-type models of whole muscles. Despite impactful applications in numerous fields, the neuromechanical information and the physiological accuracy such models provide remain limited because of multiscale simplifications that limit comprehensive description of muscle internal dynamics during contraction. We addressed this limitation by developing a novel motoneuron-driven neuromuscular model, that describes the force-generating dynamics of a population of individual motor units, each of which was described with a Hill-type actuator and controlled by a dedicated experimentally derived motoneuronal control. In forward simulation of human voluntary muscle contraction, the model transforms a vector of motoneuron spike trains decoded from high-density EMG signals into a vector of motor unit forces that sum into the predicted whole muscle force. The control of motoneurons provides comprehensive and separate descriptions of the dynamics of motor unit recruitment and discharge and decode the subject's intention. The neuromuscular model is subject-specific, muscle-specific, includes an advanced and physiological description of motor unit activation dynamics, and is validated against an experimental muscle force. Accurate force predictions were obtained when the vector of experimental neural controls was representative of the discharge activity of the complete motor unit pool. This was achieved with large and dense grids of EMG electrodes during medium-force contractions or with computational methods that physiologically estimate the discharge activity of the motor units that were not identified experimentally. This neuromuscular model advances the state-of-the-art of neuromuscular modelling, bringing together the fields of motor control and musculoskeletal modelling, and finding applications in neuromuscular control and human-machine interfacing research.

NeuroMechanics: Electrophysiological and Computational Methods to Accurately Estimate the Neural Drive to Muscles in HumansIn Vivo

“…Estimations of the neural drive at the muscle scale improved with increasing the density of surface electrode grids and the number of electrodes (as displayed in Figure 2 and demonstrated by Caillet et al (2023a)). For example, the low-density grid of 35 electrodes returned an inaccurate estimate of the neural drive (top-left subplot in Figure 3) with sudden jumps and drops over the interval of recruitment threshold where no MUs were identified, and underestimation in the regions of low forces where the low-threshold active MUs were not identified.…”

“…Conversely, MN-driven models with MU resolution may integrate parameters that have a physiological meaning and may be measured experimentally. For example, Caillet et al (2023a) proposed advanced physiological models of the MUs' activation dynamics with parameters directly identified from recent experimental data and achieved accurate estimations of force without having to perform steps of parameter calibration.…”

“…Current source-separation algorithms tend to identify the MUs with the largest MUAPs within the EMG signal (Farina & Holobar, 2016). As high-threshold MUs have the largest innervation ratios in the MU pool (see Figure 2 in Caillet et al (2023a)), decomposition techniques tend to favour the identification of the higher-threshold MUs. Therefore, by usually over-representing high-threshold over low- threshold MUs, the identified MU samples are likely not representative of the MU pool, as they do not follow an exponential distribution of recruitment thresholds (i.e., a greater proportion of low- threshold MUs over high-threshold MUs).…”

“…The data and results were obtained in Caillet et al (2023a, 2023b). EMG signals were recorded from the tibialis anterior muscle with three grid configurations of increasing electrode density (from left to right: grids of 12-mm, 8-mm, and 4-mm interelectrode distance involving 35, 64, and 256 electrodes, respectively) during trapezoidal contractions up to 30% maximum voluntary contraction (MVC) (first row) and 50% MVC (second row).…”

“…In response to a rising net excitatory synaptic drive, the MUs are recruited sequentially in order of size (Henneman’s size principle; Henneman, 1957; Henneman, 1981), starting with the smallest MUs (small MNs, low current and force thresholds for recruitment, low generated forces). Human MU pools include many low-threshold MUs and few high-threshold MUs following an exponential continuous distribution (Duchateau & Enoka, 2022; Caillet et al, 2023a). The smaller MUs usually attain higher discharge frequencies (De Luca & Hostage, 2010) and undergo a more rapid acceleration of the firing rate at recruitment and a more pronounced rate limiting than larger MUs (Avrillon et al, 2023a).…”