JMIR Publications

ABSTRACT

Background:

Wearable physiological monitoring devices are promising tools for remote monitoring and early detection of potential health changes of interest. The widespread adoption of such an approach across communities and over long periods of time will require an automated data platform for collecting, processing, and analyzing relevant health information.

Objective:

In this study, we explore prospective monitoring of individual health through an automated data collection, metrics extraction, and health anomaly analysis pipeline in free-living conditions over a continuous monitoring period of several months.

Methods:

A total of 59 participants provided smartwatch data and health symptom and illness reports daily over an 8-month window. Physiological data, including high-resolution interbeat interval (IBI), were uploaded directly from the smartwatches and processed automatically using a modular software architecture. A health risk algorithm developed in a previous influenza challenge study using electrocardiogram (ECG) sensors was applied to data collected with the wrist-worn photoplethysmography (PPG) sensors. Health risk scores exceeding a predefined threshold were flagged and checked for corresponding symptom or illness reports. From the self-reported health survey responses, illness reports of influenza and COVID-19 were noted and checked for corresponding changes in health risk score.

Results:

The median average percentage of sensor data provided per day indicating smartwatch wear compliance was 69%, and survey responses indicating health reporting compliance was 46%. A total of 29 elevated health scores were detected, of which 41% had concurrent survey data and indicated a health symptom or illness. A total of 21 influenza or COVID-19 illnesses were reported; 43% of these reports had concurrent smartwatch data, of which 67% had an increase in health score that was above or below threshold.

Conclusions:

We demonstrate a protocol for data collection, extraction of metrics, and prospective analysis that is compatible with near real-time health assessment using wearable sensors for continuous monitoring. The modular platform allows for choice of different wearable sensors and algorithms for health anomaly detection. Although data reporting compliance was at a sufficient level in general for accurate calculation of health risk scores, limited wear compliance or health survey reporting limited confirmation of illness in many cases. To our knowledge, the study demonstrates for the first time the feasibility of measuring high-resolution heart interbeat interval and step count using smartwatches in real time for illness detection over a long-term monitoring period in free-living conditions.