Penerapan Internet of Things dan Metode Gaussian Naive Bayes dalam Identifikasi Dini dan Monitoring Kondisi Jantung Secara Real-Time

Abstract

Facial Cardiovascular disease (CVD) is the leading cause of death globally, where early detection is key to effective management. Conventional episodic monitoring systems often fail to detect anomalies in a timely manner. This study aims to design and implement an end-to-end Internet of Things (IoT) system for real-time heart condition monitoring. This system integrates ESP32 microcontroller-based wearable devices with biomedical sensors (MAX30102, MLX90614, PSG010) for vital sign data collection. Data is sent in real-time to an interactive web dashboard built using React and Node.js via Firebase Realtime Database. A Gaussian Naive Bayes classification algorithm, trained using a public dataset, is implemented to classify heart conditions into three risk categories. Test results show that the system successfully achieves end-to-end monitoring with an average latency of 1.4 seconds. The performance of the Gaussian Naive Bayes model on internal test data achieves 85,35% accuracy. Additionally, the system was verified through 63 real-world functional tests, demonstrating the consistency and reliability of the prototype in measuring and classifying health data. This system prototype demonstrates the technical feasibility of a smart, affordable, and accessible heart monitoring solution that has the potential to improve proactive early detection of PKV.