Skip to main content
2024 BMES Annual Meeting Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

High School Student Local to the Baltimore Area

HS Poster

Poster 159 - Validation of Wearable Technology to Enable Health Equality Towards Accurate Data for All

Saturday, October 26, 2024
10:00 AM - 12:00 PM EST
Location: Exhibit Hall E, F & G

Introduction: Wearable devices offer a unique and non-invasive way to collect a suite of relevant biomechanical and physiological data relevant for human health and performance. Wrist-worn devices often collect data through photoplethysmography (PPG), leveraging light reflectance and absorbance to discern changes in blood volume to quantify physiological metrics like blood oxygen saturation (SpO2). Pulse oximetry technology leverages the difference in light absorption between deoxygenated and oxygenated hemoglobin to monitor oxygen saturation. This technology is considered to be a reliable alternative to the more invasive arterial blood gas measurement, which requires trained professionals and specialized analysis equipment. In parallel to the changing clinical needs for home monitoring, pulse oximetry has been integrated into multiple common consumer smartwatches. However, evidence of the performance of these devices remains limited. The few studies that do exist do not investigate the effects of individual characteristics (e.g., skin tone) and age, compare across multiple devices or thoroughly examine performance in low ranges, or explore patterns of error and bias (e.g., consistent over- or underestimation of measurements) or missingness. With the increasing number of commodity-based devices used by the everyday population with their downstream effects for health-specific applications such as lung, cardiovascular, or neuromuscular disease, there remains a critical need for the validation of such technologies to affirm their reliability against gold standard monitoring technologies across subjects of varied skin tones, BMI, gender, and age.

Materials and

Methods: In this pending Institutional Review Board (IRB)-approved study, we seek to assess the efficacy and validity of six devices for monitoring SpO2: Apple Watch 7, Garmin Fenix 7 Sapphire Solar, Welch Allyn, GE Dash 5000, Wellue ring, and Zacurate Fingertip Pulse Oximeter. The study seeks to enroll n=100 healthy individuals at Lehigh University. Upon enrollment, patients’ skin tone will be measured on the ITA scale using the Delfin SkinColorCatch, and visually on the Fitzpatrick scale. Subjects will be asked to keep still and will be in the seated or reclining position during measurement to eliminate the difference between the dominant hand and non-dominant hand. During the study procedure, if a smartwatch fails to produce a measurement, it will be readjusted and repositioned on the subject’s wrist. If the device still fails to produce a measurement, no value will be recorded for this specific time point, and it will be labeled as a missing (unobtainable) observation. Subjects will wear three wearable devices over a 10-minute interval, with a trained clinical research fellow taking readings from the devices every minute. Each subject will then do a second round of measurements with the remaining three devices over a 10-minute interval for a total duration of 20 minutes. Each subject will provide 600 data points with planned collection of 6,000 data points. Data analysis will consist of Bland Altman’s plots, scatterplots, and combined scatter plots in Python (Version 3.11.7) and Microsoft Excel (Version 2406).

Results, Conclusions, and Discussions: Preliminary and unpublished feasibility results of this study demonstrate that the Garmin Fenix 7 Sapphire Solar watch underreports SpO2 measurements by ~3-4%. Based on tangential studies evaluating the efficacy of heart rate in wearable PPG devices analogous to the above, we expect the Apple Watch and Garmin watches to underestimate or overestimate SpO2 levels with such differences being more pronounced in individuals with darker skin tones, quantified by the Delfin SkinColorCatch device. We expect the correlation of the SkinColorCatch to the Fitzpatrick scale to be satisfactory, highlighting the need for objective data to be integrated into the clinical decision-making process. When completed, this will be the largest study in the biomedical engineering community to assess the efficacy of wearable pulse oximeters in the healthy everyday population. Our data will be uploaded to open-access repositories. The Seshadri Laboratory hopes that this study can result in future academic-industry partnerships to enable the development and dissemination of validated technologies for health monitoring.

Acknowledgements (Optional): N/A

References (Optional): N/A