Personalized Analytics and Wearable Biosensor Platform for Early Detection of COVID-19 Decompensation (DECODE)

The goal of this project was to develop an artificial intelligence-based data analytics and cloud computing platform, paired with U.S. Food and Drug Administration (FDA)-cleared wearable devices, to create a personalized baseline index that could indicate a change in health status for patients who have tested COVID-19 positive. The project involved the development and validation of a COVID-19 Decompensation Index (CDI) that was built off physIQ's existing wearable biosensor-derived analytics platform. Data were collected from 1,000 (400 Phase I and 600 Phase II) human subjects that were both pre-hospitalization subjects (found to be positive for COVID-19) and subjects that had been hospitalized and treated for COVID and then discharged. This combined population consisted of COVID-19 decompensation cases (defined as ≥ 3 on the WHO COVID-19 Ordinal Scale) and cases for which COVID-19 did not result in any kind of decompensation. The Phase I data were utilized for training of a CDI algorithm while Phase II data was utilized as a validation dataset. Performance was assessed using receiver operator characteristics (ROC) area under the curve (AUC) as the metric of performance.