Desenvolvimento e validação da análise de dados de EMGS em tempo real na avaliação da fadiga muscular

Abstract

This study focuses on developing and validating real-time and offline sEMG signal analysis codes to enhance accuracy and clinical applicability in neuromuscular disorders. In the introduction, the importance of electromyography as a diagnostic and monitoring tool is emphasized, especially for neuromuscular conditions, alongside the need for automated strategies to analyze large datasets. The objective was to build, validate, and compare tools for sEMG signal analysis in two distinct contexts: real-time with the conventional method, employing literature-based approaches for preprocessing and feature extraction. In the methods, the real-time code implemented Butterworth filters, FFT calculation, and continuous analysis with CWT and was compared with the conventional method, which performs the same analyses. The results showed a high correlation between the two methods, validating the consistency of analyses and highlighting the advantages of each approach in clinical and experimental contexts. The discussion compared findings with the literature, showcasing the compatibility of the methods and innovations in hybrid approaches for signal analysis. Finally, in the conclusion, the practical applicability of the developed tools was confirmed, highlighting the future potential of integrating machine learning algorithms to optimize sEMG signal analyses and personalize clinical interventions