Algorithm for mobile analysis of ECG predicts R peaks with 99.7% sensitivity
Researchers at Technion-Israel Institute of Technology have developed and tested an algorithm within a mobile app capable of acting as an electrocardiogram (ECG) R peak detector for arrhythmogenic events. Findings were published May 22 in JMIR mHealth and uHealth.
In this study, researchers outlined the development of an algorithm to monitor R peaks and tested the system on patient data.
“An algorithm that runs in real time and precisely calculates the heart rate from electrocardiogram (ECG) signals on a beat-to-beat basis can serve as the core of a mobile system to remotely monitor patient health and issue alerts in the case of cardiac events,” wrote first author Vadim Gliner, MSc, and colleagues. “Due to their increasing computational power, wireless and connectivity, and the ability to store data on the cloud, mobile phones and tablets can run real-time algorithms to alert the patient and communicate with the medical staff.”
The study used the MATLAB mobile platform as the base to implement the algorithm, which was trained on the annotated non–atrial fibrillation MIT-BIH Arrhythmia Database. The algorithm used motion artifacts, electrical drift, breathing oscillations, electrical spikes and environmental noise to detect the R peak in electrocardiogram readings.
Results showed the MATLAB platform with implemented algorithm was able to detect R peaks with a sensitivity of 99.7 percent and positive prediction of 99.4 percent. Researchers noted these findings were of higher-quality than currently used algorithms.
“Accurate real-time identification of heart rate on a beat-to-beat basis in the presence of noise and atrial fibrillation events using a mobile phone is feasible,” concluded Gliner and colleagues. “A mobile health app with a robust R peak detector is necessary to calculate heart rate to diagnose diseases, evaluate the patient’s condition, and trigger alerts if potentially fatal events are about to occur or have just occurred. To identify real-time R peak intervals, the algorithm must deal with many kinds of common artifacts before it can be implemented on a mobile bundle.”
Cara joined TriMed Media in 2016 and is currently a Senior Writer for Clinical Innovation & Technology. Originating from Detroit, Michigan, she holds a Bachelors in Health Communications from Grand Valley State University.