Estimating Continuous Muscle Fatigue For Multi-Muscle Coordinated Exercise: A Pilot Study on Walking

TL;DR

This work tackles the challenge of continuously estimating muscle fatigue during daily, multi-muscle, submaximal walking tasks. It introduces physiology-inspired features from muscle synergy fractionation and spinal module activation, and fuses them within a Bayesian Gaussian process framework trained with a monotonicity-based loss to capture time-evolving fatigue without explicit labels. The approach demonstrates cross-day and cross-subject stability and shows superior trendability and suitability compared with baseline estimators and alternative views of fatigue. The results suggest practical utility for rehabilitation planning, personalized dosing, and human-machine interfaces in contexts like the Metaverse, where continuous fatigue monitoring is valuable.

Abstract

Assessing the progression of muscle fatigue for daily exercises provides vital indicators for precise rehabilitation, personalized training dose, especially under the context of Metaverse. Assessing fatigue of multi-muscle coordination-involved daily exercises requires the neuromuscular features that represent the fatigue-induced characteristics of spatiotemporal adaptions of multiple muscles and the estimator that captures the time-evolving progression of fatigue. In this paper, we propose to depict fatigue by the features of muscle compensation and spinal module activation changes and estimate continuous fatigue by a physiological rationale model. First, we extract muscle synergy fractionation and the variance of spinal module spikings as features inspired by the prior of fatigue-induced neuromuscular adaptations. Second, we treat the features as observations and develop a Bayesian Gaussian process to capture the time-evolving progression. Third, we solve the issue of lacking supervision information by mathematically formulating the time-evolving characteristics of fatigue as the loss function. Finally, we adapt the metrics that follow the physiological principles of fatigue to quantitatively evaluate the performance. Our extensive experiments present a 0.99 similarity between days, a over 0.7 similarity with other views of fatigue and a nearly 1 weak monotonicity, which outperform other methods. This study would aim the objective assessment of muscle fatigue.

Figures (12)

• Figure 1: The schematic plot of the workflow. The exercise variable $\bm{x_t}$ is extracted from IMU measurements of shank and thigh. The fatigue-induced spatial and temporal patterns of multiple muscles $W_t$ and $I_t$ are extracted from EMG signals.Then, a directed graph model is built to infer the latent fatigue given the observations, Then, a physiology-inspired loss function is used to train the algorithm.
• Figure 2: The schematic plot of the experimental paradigm. The whole experiment consists of three sessions, each lasting 20 minutes (dark grey) and followed by data storage periods (blue). For each session, the subjects are asked to walk with the speed of 4.5 km/h and the question of fatigue feelings is asked every 5 minutes.
• Figure 3: The schematic plot of fatigue dynamic model, corresponding to how the hidden state muscle fatigue $f_t$ induces the muscle compensation $W_t$ and the variation of spinal module spike timings $I_t$.
• Figure 4: The schematic plot of $y=\Delta(x)$.
• Figure 5: Fatigue Score Estimation