Parkinson’s disease is a progressive neurological disorder that affects movement, muscle control, and coordination, making early detection and continuous monitoring critical. Traditional diagnosis often depends on clinical observation, which may fail to capture subtle or early-stage symptoms. This project proposes an AI and IoTbased system that utilizes wearable sensors to monitor physiological and motion-related parameters, including heart rate, muscle activity (EMG), and hand tremors (MEMS). The system integrates these sensors with a NodeMCU microcontroller to collect real-time data, which is subsequently processed using a Python-based backend. An XGBoost machine learning model analyzes the synchronized data to predict the likelihood of Parkinson’s disease. Results are visualized through a Streamlit dashboard, while a buzzer and LCD provide instant local alerts. This low-cost, portable solution aims to support early detection, continuous monitoring, and enhanced healthcare assistance for Parkinson’s patients.
Panchetti et al. (Wed,) studied this question.