Integrated Portable ECG Monitoring System with CNN Classification for Early Arrhythmia Detection

Modigari Narendra ^1* Aayush Panwar ² Arnav Arya ² Rohan Raj ² Arnab Kumar ²

• ¹ School of Computer Science and Engineering, Vellore Institute of Technology (VIT), Chennai, India
• ² School of Electronics Engineering, Vellore Institute of Technology (VIT), Chennai, India

The electrocardiogram (ECG) is vital for diagnosing heart diseases, offering insights into cardiac health through electrical activity patterns. Portable monitoring solutions have emerged to meet the growing demand for timely evaluations. This paper introduces a portable ECG monitoring system featuring classification capabilities using Convolutional Neural Networks (CNN). An Arduino Nano microcontroller and an AD8232 ECG sensor are used for cost-effective, continuous monitoring. The hardware involves reliable ECG signal acquisition by interfacing the AD8232 sensor with the Arduino Nano. CNNs enhance real time ECG data classification, enabling identification of cardiac abnormalities, including arrhythmias. Extensive training on MIT-BIH datasets illustrates the CNN model's accuracy. It achieves binary and multi-class classification accuracies of 98.35% and 99.3%, respectively, surpassing existing benchmarks. The system's strong performance highlights its potential impact on clinical practice, providing a cost-effective and reliable solution for continuous ECG monitoring. The proposed portable ECG monitoring system with CNN-based classification shows great promise for early arrhythmia diagnosis and continuous cardiac monitoring. Its accurate classification skills, low-cost hardware, and real-time monitoring features make it an effective tool for improving cardiovascular health management.