Fereidoun Nowshiravan Rahatabad - Department of Biomedical Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran.
Keivan Maghooli - Department of Biomedical Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran
Talat Rahimi Balan - Department of Biomedical Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran

Introduction: “Muscular synergy” is one of the methods for determining the relationship between the central nervous system and muscles which are involved in performing a specific movement. To perform each movement, certain patterns are followed through the central nervous system that control the number of synergies, and these patterns are modified and optimized during the skill. Thepresent study aimed to classify basketball athletes based on muscular synergy analysis. Material and Methods: For the purpose of the study, the electromyography (EMG) signals of six dominant hand muscles were recorded during performing three basketball skills. Subsequently, synergy was identified using the non-negative matrix factorization method. In the next stage, the cosine similarity feature was calculated separately; furthermore, the time and frequency of the main signal were analyzed, and the neural network was evaluated using MATLAB software. Results: The result of the classification was obtained by applying the dimensioned reduced matrix of all the existing features with a reliability of 73.68%. In addition, the results demonstrated that the cosine similarities between the muscles of each person could lead to the training of the neural network and classification of individuals at different levels of skill. Conclusion: The present study suggested a new idea regarding synergistic features for classifying athletes based on EMG signal. However, the use of time and frequency features only facilitated differentiation between a maximum of two groups.

Neural Networks, Computer, Performance, Sports