Wearable monitoring of COVID-19 biomarkers

T has reported an automated wireless wearable sensor and a machine-learning approach that records and interprets mechano-acoustic signatures of a person's heart rate, respiration rate, and temperature, along with coughing, speaking, and laughing, all of which are biomarkers relevant to monitoring the progression of COVID-19 infections in individuals and their spread across populations.

he researchers at Proceedings of the National Academy of Sciences of the United States of America (PNAS)

Results from 37 people diagnosed with COVID-19 revealed patterns of evolution of these parameters during all stages of the disease, from initial diagnosis through hospital treatment and final recovery at home.

Systematic imaging studies demonstrate correlations between the time and intensity of coughing, speaking, and laughing and the total droplet production, as an approximate indicator of the probability for disease spread. The sensors, deployed on COVID-19 patients along with healthy controls in both inpatient and home settings, record coughing frequency, and intensity continuously, along with a collection of other biometrics.

The results indicate a decaying trend of coughing frequency and intensity through the course of disease recovery, but with wide variations across patient populations. The methodology creates opportunities to study patterns in biometrics across individuals and among different demographic groups.

As COVID-19 is a respiratory disease, cough and other sounds from the thoracic cavity, trachea, and esophagus are examples of highly relevant biometrics. Laboratory-scale studies demonstrate cough-based diagnoses of diverse respiratory diseases through measurements of frequency, intensity, persistency, and unique audio features. Investigations on audio recording data show differences between COVID-19 positive and negative subjects' vocalizing patterns including phonation of speech, breathing, and coughing sounds. The results may suggest possibilities for disease monitoring in asymptomatic patients.

The results presented here bypass these disadvantages, to allow continuous assessments of respiratory biomarkers correlative to health status and droplet/aerosol production, with additional information on a range of traditional vital signs.

Here, a simple, wireless monitoring device combines with a cloud interface and a data analytics approach to allow continuous monitoring of a breadth of conventional (e.g., heart rate, respiratory rate, physical activity, body orientation, and temperature) and unconventional (e.g., coughing, speaking) physiological parameters of direct relevance to COVID-19.

The results serve as a quantitative basis for 1) detecting early signs of symptoms in health care workers and other high-risk populations, 2) monitoring symptomatic progression of infected individuals, and 3) tracking responses to therapeutics in clinical settings. In addition, systematic studies presented here indicate that coughing, speaking, and laughing events measured with these devices correlate to the total amount of droplet production.

This link offers an opportunity to quantify the infectiousness of individuals, as critical information in caring for patients and for improved risk stratification in the context of contact tracing and individual quarantines.

The statistics may provide insights for creating guidelines for disease management and containment. Further studies on an expanded patient population with detailed demographic information are, however, necessary to enable big-data-based studies of the demographic dependence and/or individual variance of relevant biometrics.

( With inputs from ANI )