Frontiers in Computer Science (Feb 2024)
HRVEST: a novel data solution for using wearable smart technology to measure physiologic stress variables during a randomized clinical trial
Abstract
The purpose of this study was to address the logistical and data challenges of using wearable technologies in the context of a clinical trial to measure heart rate variability (HRV) as a marker of physiologic stress in emergency healthcare providers during the COVID-19 pandemic. When using these wearable smart garments, the dilemma is two-fold: (1) the volume of raw physiological data produced is enormous and is recorded in formats not easily portable in standard analytic software, and (2) the commensurate data analysis often requires proprietary software. Our team iteratively developed a novel algorithm called HRVEST that can successfully process enormous volumes of physiologic raw data generated by wearable smart garments and meet the specific needs of HRV analyses. HRVEST is a noise-filtering and data-processing algorithm that allows the precise measurements of heart rate variability (HRV) of clinicians working in an Emergency Department (ED). HRVEST automatically processed the biometric data derived from 413 electrocardiogram (ECG) recordings in just over 15 min. Furthermore, throughout this study, we identified unique challenges of working with these technologies and proposed solutions that may facilitate future use in broader contexts. With HRVEST, using wearable smart garments to monitor HRV over long periods of time becomes logistically and feasibly viable for future studies. We also see the potential for real-time feedback to prophylactically reduce emergency physician stress, like informing optimal break-taking or short meditation sessions to lower heart rate. This could improve emotional wellbeing and, subsequently, clinical decision-making and patient outcomes.
Keywords
