Abstract: Chronic diseases like diabetes, cardiovascular, and respiratory conditions require continuous monitoring, which traditional healthcare systems often fail to provide. To address this, the proposed system introduces an AI-powered, IoT-integrated health monitoring solution for real-time, remote tracking. A Raspberry Pi 4 acts as the processing unit, receiving data from a Near-Infrared (NIR) sensor managed by an ESP8266 microcontroller. This setup allows for non-invasive monitoring of key health metrics such as Haemoglobin A1c, insulin levels, caloric intake, and lung function indicators. The system uses a Long Short-Term Memory (LSTM) neural network to analyse time-series data and predict health risks. It classifies a patient’s condition as normal or at risk, enabling early detection and timely intervention. Continuous monitoring reduces hospital visits and empowers patients to manage their health independently. Healthcare providers receive real-time, actionable insights for better decision-making. The integration of AI and IoT ensures accurate data collection and intelligent analysis. Overall, the system supports a proactive, patient-focused approach to chronic disease management.
Keywords: AI-based algorithms, diabetes management, esp8266 module, non-invasive monitoring, NIR sensor, raspberry pi 4b, LSTM neural network, remote health monitoring, wireless data transmission, sensor-based systems.
|
DOI:
10.17148/IARJSET.2025.12456