Leveraging Smart Bed Technology to Detect Symptoms of Influenza-Like Illnesses: A Case Study Focused on COVID-19
ABSTRACT
Background:
Pathophysiological responses to viral infections such as COVID-19 affect sleep duration, sleep quality, and concomitant cardiorespiratory function. The collection of real-world, longitudinal sleep-metric data could prove to be valuable for infectious illness detection and monitoring.
Objective:
The aim of this study was to leverage longitudinal, retrospective, biometric data captured using ballistocardiography (BCG) signals from a consumer smart bed platform along with predictive modeling to detect and monitor, at an individual level, COVID-19 symptoms.
Methods:
In this study, objective sleep metrics were obtained using a smart bed platform from survey respondents (N = 1725) reporting the outcome of a COVID-19 test. Using these sleep metrics, we developed a two-stage model detecting the presence of symptoms and the progression of illness associated with COVID-19. First, a gradient-boosted, decision-tree, “symptom detection model” was created to classify each sleep session as symptomatic or not. Second, an “influenza-like illness-symptom progression model” utilized a Gaussian Mixture Hidden Markov Model to estimate the probability of each sleep session to be symptomatic or not by leveraging the temporal dimension of the collected data.
Results:
Symptoms were detected in 104 of the 122 positive COVID-19 cases. Group level trends indicated the model could detect differences in sleep metrics during symptomatic versus reference periods. The versatility of our model extends beyond COVID-19 and can be applied to a range of respiratory illnesses.
Conclusions:
The sleep metrics measured with a smart bed platform are a unique source of longitudinal data, collected in a real-world and unobtrusive manner. In the future, this system may serve as an asset in predicting and tracking the development of symptoms associated with a wide variety of respiratory illnesses. Clinical Trial: The protocol describing this observational study # P2020/03001 NEIRB 17-1323071-1 was approved by the WCG New England Institutional Review Board (IRB).
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.