A wearable ECG monitoring system for real-time arrhythmia detection

• Published in: 2018 IEEE 61st International Midwest Symposium on Circuits and Systems (MWSCAS) pp. 787 - 790
• Main Authors: Clark, Nicholar, Sandor, Edward, Walden, Calvin, In Soo Ahn, Yufeng Lu
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2018
• Subjects: arrhythmia; Biomedical monitoring; Detection algorithms; DSP; ECG monitoring; Electrocardiography; Hardware; Monitoring; real-time; Real-time systems; Signal processing algorithms; wearable
• DOI: 10.1109/MWSCAS.2018.8624097

This study develops a wearable medical system for real-time arrhythmia detection, which acquires electrocardiogram (ECG) data through a three-lead ECG sensor. It performs ECG signal processing and immediately alerts the patient's health care provider of an arrhythmia via wireless messaging. In particular, a common form of arrhythmia known as premature ventricular contractions (PVCs) are identified using the Pan-Tompkins and the wavelet-based template-matching algorithms. When three or more consecutive PVCs are detected, the device sends urgent report email to a patient's health care provider. In the experimental study, the design has been successfully validated using benchmark records from the MIT-BIH arrhythmia database. A low-cost Texas Instruments TMS320C5515 and Raspberry Pi 3 Model B are chosen to be the hardware platform. This study suggests a viable, low-complexity solution for real-time heart monitoring and arrhythmia detection.