Feasibility and reliability of whintings scanwatch to record 4-lead Electrocardiogram: A comparative analysis with a standard ECG

Abstract

The recent technological advances allowed us to produce some new wearable devices, such as smartphones and smartwatches (SW). These devices provide different services to their users through different software applications installed even in the SW or smartphones. Health monitoring functionalities, among these services, are nowadays the new technological fashion. In fact, the monitoring is ensured by the sensor incorporated in the SW. The SW allows the record of only one single lead Electrocardiogram (ECG), which is sufficient to screen or diagnosis of rhythm and conduction disorders, especially during the onset of cardiac symptoms, but insufficient for the detection of ischemic disease and cardiomyopathies. In this context, this paper aims to evaluate the feasibility, and reliability of a SW to obtain ECG recordings in comparison with a standard ECG. For that purpose, 140 patients were recruited for this analysis. At the first step, the 12 lead ECG followed with four lead SW-ECG; using the Withings Scanwatch device, were recorded in the same resting conditions. The four lead SW-ECG consists of Einthoven DI lead recorded with the SW, where the SW was on the left wrist and the right index finger on the crown, and three Wilson-type leads, in the which the V1 was recorded in the fourth right parasternal intercostal space, V3 was recorded in the fifth intercostal space on the midclavicular line, and V6 was recorded in the fifth intercostal space on the left midaxillary line with the right index finger placed on the crown and the left hand encompassing the right wrist. 700 ECGs recordings were collected and statistically analyzed in this study. In total, 97 % of the patients were able to obtain an ECG through the SW. A strong correlation was observed between the two recording methods concerning the duration of the studied parameters (r >90 %). The correlation coefficient showed that 33 out of 44 parameters have a strong correlation with the standard ECG. The similarity of the combined leads in the 4 established subgroups was significantly higher, meaning that increasing the number of leads would improve the detection of electrical anomalies. Our findings confirm the existing data on the high similarity between SW and standard 12-leads ECG. Despite SW not having the accuracy and utility of the standard ECG machine, they should be considered as an interesting screening tool for cardiac rhythm disorders, and a compelling solution to electrical documentation of general cardiac symptoms.