A Muscle Fiber Conduction Velocity Estimator Using Surface Electromyography Signals Acquired From Vastus Lateralis

Muscle fiber conduction velocity (MFCV) is an efficient tool for analyzing fatigue and underlying myoneural abnormalities. It provides important information about the motor unit recruitment strategies better than conventional spectral variables during various muscle activities. In this work, an attempt has been done to design and implement an MFCV estimator using surface electromyography (sEMG). Signals are acquired from muscle Vastus lateralis (VL) during isometric knee extension using a well-defined protocol. Further, they are filtered and processed using Raspberry Pi Model B+. First and the last two seconds of the processed signal denotes Non-fatigue (NF) and Fatigue (F) respectively. MFCV is calculated by determining the propagation delay using the autocorrelation algorithm that requires only a single data acquisition channel. In order to validate the fatigue status of the muscle, a spectral feature named Mean frequency (MNF) is also extracted. The threshold values for both MFCV and MNF are set after analyzing the variation of these features over the entire contraction time. After comparing both the features of the input signal with their corresponding threshold, the muscle condition is identified as NF or F. A decrease of 12.3% is observed in MFCV during F when compared to NF. A similar trend is seen in MNF with a reduction of 24.7% which confirms the fatigue state of the muscle. The status of the muscle with the values of MFCV and MNF are displayed. This device is connected to Telegram mobile application using IoT. The obtained MFCV estimates are validated with the range reported in the literature. Hence, the designed MFCV estimator may be used for fatigue assessment in the field of sports science and biomedicine. The proposed device might be further modified to identify the underlying fiber type dominance in different muscles.