The quantification of performance-determining factors is an important and fundamental aspect in the training process. Here, coaches and athletes alike are given a feedback on the current performance status and development.
Current methods for determining the training state are, e.g. the analysis of blood lactate or the analysis of the composition of the respiratory gas (spirometry). The parameters extracted with these methods are among the standards of sports medicine and correlate reliably with physiological thresholds. Thus, information is available for trainers and athlete about individual optimal training zones. The determination of these parameters, however, is very complex and gives only a feedback on the global state of the athletes. Electromyography (EMG) detects the myo-electrical activities of individual muscles. The signal from the superficial muscles is derived from the surface of the skin and equal to the sum of the muscle fiber's activity, which are located underneath the electrode. A myo-electrical signal already contains very fundamental information about the characteristics of a muscle and gives a first essential understanding of the neuromuscular activities. The often unknown and individual composition of the muscle from types I, IIa and IIb fibers leads to the assumption that the muscles react individually under specific loads. It is known that during anaerobic endurance stresses, the fatigue-resistant type I fibers are predominantly active. As the load increases in the sense of higher motion resistance or prolonged load time, the muscle and the ratio of active fibers from type I and type II a, b is changed.
Since the muscle fibers also have different myo-electrical properties, this change also leads to a change in the power spectral density of the myo-electrical signal. Numerous studies already confirm the existence of these so-called EMG thresholds. In this case, the athlete could benefit from the analysis of local fatigue and thus specifically identify and train the weakest muscle, or consciously initiate compensation strategies. These are conclusions that are not possible with current performance-diagnostic approaches. However, the reliability with which these so-called EMG thresholds are determined is currently not sufficient. EMG specific signal processing steps are of particular importance in this regard in order to confirm the reliability and validity of the findings. In addition, novel methods are used to safely detect active muscular activity and to extract additional features for performance diagnosis. For this purpose, the myoelectric signals are analyzed in the time domain, in the frequency domain and in the time-frequency domain. A detailed analysis of the EMG signals, could help in an early prediction of muscle fatigue.
The aim of this interdisciplinary engagement is the reliable quantification of load, level- and fatigue-dependent changes in the EMG signal up to an integration of myo-feedback into the training process of performance and broad sports. In cooperation of the Chair for Signal Processing and System Theory (Gerhard Schmidt) and the Institute of Sports Science (Stefan Kratzenstein), innovative methods of digital signal processing are combined with physiological processes in the course of movement.
S. Kratzenstein, E. Elzenheimer, E. Suresh, G. Schmidt: Automatische Verfahren zur Bestimmung des neuromuskulären Aktivitätsbeginns in Anhängigkeit von der Signalqualität, GAMMA, Hamburg, Germany