A portable real-time ECG recognition system based on smartphone

Area of Science:

• Biomedical Engineering
• Signal Processing
• Medical Informatics

Background:

• Real-time monitoring of electrocardiogram (ECG) signals is crucial for diagnosing cardiac conditions.
• Existing systems often face challenges with portability, real-time processing, and accuracy in complex environments.
• Development of efficient computer-aided diagnosis (CAD) systems for cardiology is an ongoing research area.

Purpose of the Study:

• To develop and evaluate a smartphone-based system for real-time ECG monitoring and heartbeat recognition.
• To implement a noise-tolerant algorithm for accurate identification of various heartbeat types.
• To assess the system's performance in terms of accuracy, speed, and portability for potential clinical application.

Main Methods:

• ECG signals were acquired using a low-power microprocessor (MSP430FG4618) and transmitted wirelessly to a smartphone via Bluetooth.
• A noise-tolerant ECG heartbeat recognition algorithm utilizing discrete wavelet transform and higher-order statistics was developed.
• The system was tested for its accuracy in classifying seven different types of heartbeats and its real-time processing capability.

Main Results:

• The system achieved a high accuracy of 98.34% in identifying seven distinct types of heartbeats.
• Real-time heartbeat recognition was achieved, with an average identification time of 78 milliseconds per heartbeat.
• The developed algorithm demonstrated superior performance compared to other existing methods in the literature.

Conclusions:

• The proposed smartphone-based system offers a portable and effective solution for real-time ECG monitoring and heartbeat recognition.
• The high accuracy and rapid processing speed highlight the system's potential as a practical computer-aided diagnosis (CAD) tool.
• This technology could significantly enhance the accessibility and efficiency of cardiac diagnostics.