Abstract:
This is an advanced heart attack detection system based on real-time MPU-6050 (motion tracking), AD8232 (ECG monitoring), and MAX3010 (pulse-oximetry, heart rate) base sensor connected to pi pico and forward data to Raspberry Pi 4 (8GB) for edge computing. The system has developed a durable training approach for AI models in a collaborative work-based environment over distributive devices without compromising the privacy of the data. The model that was trained, installed upon the Pi 4, processes readings from the sensor locally and is able to predict the risks of heart attack accurately and display the results in an easy-to-understand user interface. Furthermore, the system has adopted Wazuh, which is a SIEM (Security Information and Event Management) that allows observing and securing the infrastructure against the cyber threats for the maintenance of the integrity of the data and the reliability of the system. Through multisensorfusion, federated learning, and efficient security systems, this project delivers a scalable, privacypreserving, and secure option for detecting the impending heart attack in a manner that is reasonable in both individual and clinical settings. The system will be modular, and this will allow further development with the addition of the sensors or better AI models or designing a cell-phone based interface such that it becomes a versatile tool for remote health monitoring.
