Desenvolvimento e validação de um sistema para análise neuromuscular em tempo real para pessoas com Esclerose Lateral Amiotrófica

Abstract

Introduction: Amyotrophic Lateral Sclerosis (ALS) is a severe neurodegenerative disease that progressively compromises patients’ functionality and quality of life. The analysis of surface electromyography (SEMG) and accelerometry (ACC) signals plays a fundamental role in early diagnosis and therapeutic planning. However, the high cost of devices and the complexity of neuromuscular pattern classification limit access to these technologies, especially within the context of the Brazilian Unified Health System. Given this scenario, the development of accessible and efficient systems for ALS diagnosis and monitoring becomes essential. Objectives: This project aims to develop and validate a system for the real-time collection, processing, and analysis of EMG and ACC signals. The tool is designed to assist in the screening of ALS patients, accelerating the diagnostic process, supporting exercise prescription, and monitoring ALS progression. Methodology: The developed system consists of a low-cost SEMG sensor integrated into a computer via MQTT communication. The collected data is processed in real time, allowing instant analysis of neuromuscular signals. The validation of the sensor and processing code was conducted with a control group of four healthy individuals, ensuring the system’s reliability before clinical application. Results: The system validation included bench tests, pilot studies with four healthy individuals, and real-time processing code evaluation. EMG data were analyzed using statistical tests, including the intraclass correlation coefficient (ICC), Bland-Altman analysis, and energy ratio, demonstrating high consistency and agreement between methods. To validate the real-time processing code, raw and processed signals from each individual were compared using Pearson correlation, linear regression, and the SPM1D test, showing a strong linear correlation (r > 0.96), no statistically significant differences (p > 0.05), and regression coefficients close to 1, with high determination coefficients (R2 > 0.92). Conclusion: The developed system integrates advanced monitoring and signal analysis technologies, providing an accessible solution to support ALS diagnosis, track disease progression, and establish personalized therapeutic goals. Its high performance in preliminary validation reinforces its potential for clinical application and future incorporation into Brazilian Unified Health System.