Nontraditional Electrocardiogram and Algorithms for Inconspicuous In-Home Monitoring: Comparative Study
Casana has been devoted to the study and development of effortless, noninvasive, in-home monitoring since our very conception. The feasibility of Casana’s first product, The Heart Seat, has been tested in a number of studies, one of which is a study comparing the single-lead ECG measured by the seat to a clinical-grade 12-lead ECG. Nontraditional Electrocardiogram and Algorithms for Inconspicuous In-Home Monitoring: Comparative Study was published in a peer-reviewed journal and was authored by our Chief Scientific Officer, Nicholas J Conn PhD, our Head of R&D David A Borkholder PhD, and Karl Q Schwarz MD (affiliated with University of Rochester Medical Center).
Background: Similar to many wearables and connected in-home medical devices, The Heart Seat too will be utilized in an uncontrolled environment and measuring physiological signals at suboptimal locations on the body. The result is a reduced signal-to-noise ratio with highly variable signal quality that changes from moment to monet. The importance of sophisticated signal quality classification algorithms and robust feature delineation algorithms are required to achieve high accuracy on these poor quality physiological signals.
Creating a solution that could address patient adherence, work in an uncontrolled environment, and address concerns such as clinical validity were among the motivating factors for pursuing research on what is now known as The Heart Seat. During his PhD at RIT, Nick Conn, built custom signal quality classification algorithms to reject poor-quality regions of the electrocardiogram (ECG). A key innovation of this work was tuning the signal quality classification algorithms to assess the performance of a custom beat detection algorithm. By doing so, the good quality regions that then undergo analysis are expected to perform well, significantly reducing the potential for incorrect results. These algorithms were benchmarked against a large standard database of ECGs consisting of 900,059 heart beats and then stress tested on ECG data captured from a standard 12-lead clinical grade ECG and compared to ECG data captured from a toilet seat “buttocks ECG”, from an early rendition of The Heart Seat.
“The present algorithms were validated using a study of 25 normative subjects and 29 heart failure (HF) subjects. Measurements captured from a toilet seat-based buttocks electrocardiogram were compared with a simultaneously captured 12-lead clinical grade ECG. The ECG lead with the highest morphological correlation to buttocks electrocardiogram was used to determine the accuracy of the heart rate (HR), heart rate variability (HRV), QRS duration, and the corrected QT interval (QTc). These algorithms were benchmarked using the MIT-BIH Arrhythmia Database (MITDB) and European ST-T Database (EDB), which are standardized databases commonly used to test QRS detection algorithms.” (Conn, Schwarz, & Borkholder, 2018)
“Clinical grade accuracy was achieved for all buttocks electrocardiogram measures compared with standard Lead II. For the normative cohort, the mean was −0.0 (SD 0.3) bpm (N=141 recordings) for HR accuracy and −1.0 (SD 3.4) ms for HRV (N=135). The QRS duration and the QTc interval had an accuracy of −0.5 (SD 6.6) ms (N=85) and 14.5 (SD 11.1) ms (N=85), respectively. In the HF cohort, the accuracy for HR, HRV, QRS duration, and QTc interval was 0.0 (SD 0.3) bpm (N=109), −6.6 (SD 13.2) ms (N=99), 2.9 (SD 11.5) ms (N=59), and 11.2 (SD 19.1) ms (N=58), respectively. When tested on MITDB and EDB, the algorithms presented herein had an overall sensitivity and positive predictive value of over 99.82% (N=900,059 total beats), which is comparable to best in-class algorithms tuned specifically for use with these databases.” (Conn, Schwarz, & Borkholder, 2018)
The results of the study were overall positive. Within the subjects tested, clinical grade accuracy was achieved in comparison to standard ECG measurements when utilizing our custom signal quality classification and beat detection algorithms. Furthermore, this early validation data gave Nick et al confidence that wearable and connected devices have the potential to enable clinical-grade monitoring, providing new insights through daily, inconspicuous in-home monitoring.
To learn more about this groundbreaking work please read Nontraditional Electrocardiogram and Algorithms for Inconspicuous In-Home Monitoring: Comparative Study at JMIR Mhealth and Uhealth.
Nontraditional Electrocardiogram and Algorithms for Inconspicuous In-Home Monitoring: Comparative Study is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium.