NeuroMotion: AI‑Enabled Activity Tracking for Stroke Rehabilitation Using IoT

Stroke rehabilitation requires continuous monitoring of patient movements to evaluate recovery progress and ensure the effectiveness of therapeutic exercises. However, conventional rehabilitation methods often rely on periodic clinical assessments, which limit continuous observation and may delay the detection of improper movement patterns. To address this limitation, this paper proposes an Internet of Things (IoT)-based activity tracking system designed to monitor upper-limb movements of stroke patients during rehabilitation exercises in real time. The proposed system integrates wearable motion sensors, including an accelerometer and gyroscope, connected to an ESP32 microcontroller to capture detailed motion parameters such as orientation, velocity, and angular displacement of the patient’s arm. The sensor data is continuously collected and transmitted through Wi-Fi connectivity to a cloud-based database for storage and processing. A machine learning-based analysis module is employed to identify movement patterns, classify rehabilitation exercises, and evaluate the accuracy and consistency of patient movements.The processed data is presented through an interactive web dashboard that enables doctors, physiotherapists, and caregivers to remotely monitor patient activity and rehabilitation progress. The dashboard provides real-time visualization, historical data tracking, and performance metrics that assist healthcare professionals in making informed clinical decisions. Additionally, the system can detect irregular or incorrect movements and provide feedback for improving exercise quality. The proposed IoT-enabled rehabilitation monitoring framework offers a low-cost, scalable, and accessible solution for remote healthcare monitoring. By enabling continuous data collection and intelligent analysis of patient movements, the system enhances rehabilitation effectiveness, supports personalized therapy plans, and reduces the need for frequent hospital visits. Experimental evaluation demonstrates the feasibility of the system in accurately tracking patient activity and providing reliable data for clinical assessment.