Clinical Validation of Affordable, Reusable Novel EKG Monitor for Children in Low-Resource Settings
2Department of Pediatrics, Carilion Clinic, USA
3Department of Biomedical Engineering and Mechanics, Virginia Polytechnic Institute, USA
4Cardinal Mechtronics, USA
Background: Low-income countries severely lack adequate access to devices that accurately measure vital signs which may affect mortality outcome for pediatric patients. Developing an efficient and affordable devices that can accurately determine vital signs can contribute greatly to diagnosing critical pediatric conditions, leading to faster time to appropriate treatment. A team of engineers, industrial designers, and medical professionals developed an affordable, reusable EKG monitor prototype using carbon electrodes.
Objective: The objective of study is to validate the prototype’s accuracy compared to a standard EKG monitor.
Methods: 21 healthy subjects, ages newborn to 60 months, were recruited from Carilion Children’s General Pediatrics Clinic. Prototype device and standard EKG device were simultaneously attached to subject for 1-2 minutes. Bland-Altman method was used to evaluate if agreement existed between both devices with heartbeats counted obtained per 10-second interval (hpi) as N=1.
Results: Prototype was able to detect subject’s heartbeat and display distinct EKG waveforms. Loss of EKG detection from prototype was observed during loss of skin contact with carbon electrodes from movement![Figure 1. Bland-Altman plot comparing the difference in heartbeats counted per 10s-interval between prototype and standard EKG (blue dot). Mean of the differences was -7.5hpi (red line). Upper and lower limits of agreement (LOA) determined by ±1.96σ (Upper LOA = 8.49hpi, 95% CI [6.13,10.85]; Lower LOA = -23.5hpi, 95% CI [-25.86, -21.14]). Although all data points are between LOAs, the large range between LOAs demonstrates lack of agreement between current prototype and standard EKG.](https://events.eventact.com/dan/31002/EKG device bland altman - Copy 1.png){kind=small}
. Mean of differences between the two methods clinically translates to signficant difference of 45 heartbeats per minute. Range between limits of agreement demonstrates clinically significant discrepancy in prototype’s ability to consistently detect heart rate, presumably from loss of skin contact. However, 35% of data show difference between -2.2 to 0.4hpi between the two methods, indicating the prototype can detect heart rate similarly to the standard EKG in presumed instances of constant skin contact.
![Figure 1. Bland-Altman plot comparing the difference in heartbeats counted per 10s-interval between prototype and standard EKG (blue dot). Mean of the differences was -7.5hpi (red line). Upper and lower limits of agreement (LOA) determined by ±1.96σ (Upper LOA = 8.49hpi, 95% CI [6.13,10.85]; Lower LOA = -23.5hpi, 95% CI [-25.86, -21.14]). Although all data points are between LOAs, the large range between LOAs demonstrates lack of agreement between current prototype and standard EKG.](https://events.eventact.com/dan/31002/Bland Altman figure2.PNG)
Conclusion: For our prototype, we were able to develop a low-cost, reusable device that can detect heart rate in children and reproduce similar heart rates as the standard EKG monitor, but inconsistently, presumably secondary to loss of skin contact with the carbon electrodes. Improvements to optimize carbon electrode contact with skin will be considered for future design.
Powered by Eventact EMS