Automating the testing of athletes for muscle fatigue asymmetry

In this study, we would like to present the results of the process of developing a research method for analyzing the performance of athletes’ muscles, and more specifically, their asymmetry of fatigue[1,2]. The purpose of our system is to learn about the stronger and weaker aspects of the musculoskeletal apparatus in a group of athletes of different sports, and more specifically to detect the muscular imbalance of individual muscle parts. In addition, it was assumed to use the results of the study to draw up an individualized training plan supported by the current level of training. An example of such conduct is presented in the article [3].  Another goal was to determine the fatigue thresholds of individual muscle groups. The project is mainly focused on the musculature of the lower extremities and lower torso.

During the study, the greatest attention has been given to the areas of the following muscles: quadriceps of the thigh, biceps of the thigh, gastrocnemius, and rectus abdominis muscle. To measure the signals, an EMG system was used – the Noraxon TeleMyo 2400 G2 measuring the voltage produced by the muscle belly [4]. It consisted of eight reference electrodes collecting noise signals from the skin surface and 16 electrodes measuring electric potential generated by muscle as well as a receiving station connected to a computer. Exercises could be modified according to the subject’s needs, but the basic one were squats with a load of ten kilograms, performed in five series for twenty seconds.

The pure signal from the survey is not enough, and in order to read meaningful data from it, it must be processed properly [5]. In order to automate this process, an algorithm was created in Python language. The aim of the program is to automaticially generate numericial data and their graphical representations. Examples of the parameters were:

• the average value in the time intervals in each series,

• the average value in the series,

• the Fourier transform of the signal,

• the maximum percentage value in the series,

• the integral determined from a graph of the voltage waveform over time.

A diagram of the whole procedure can be seen in Figure 1.