SISTEM MONITORING SINYAL EMG PADA ROBOT REHABILITASI LENGAN BAGI PENDERITA STROKE BERBASIS WEBSITE

Abstract

This study discusses the design of a monitoring system based on electromyography (EMG) signals integrated with an arm rehabilitation robot. The system is designed to support patient condition analysis through a website, which serves as a platform for monitoring, therapy decision-making, and indirect control of the robot. Patient EMG data are obtained through sensors and processed by an ESP32 microcontroller to generate parameters such as Mean Frequency (MNF), Median Frequency (MDF), Mean Power (MPw), and Average Power (APw). These parameters are transmitted to the Firebase Realtime Database for website access. The website displays patient data in tables and graphs, and it is equipped with an artificial intelligence (AI) module to classify muscle strength levels based on the Manual Muscle Testing (MMT) scale. Based on this classification, the system provides exercise recommendations that can be reviewed and adjusted by doctors before being sent to the robot. Testing shows that the system workflow, from data acquisition to instruction delivery, operates consistently with an average response time of 1–2 seconds. Thus, this system can serve as a web-based monitoring solution that supports a more structured and well-documented patient rehabilitation process.