Originally published in Medical Device Network
The National Institutes of Health (NIH) unit National Cancer Institute (NCI) and the National Institute of Biomedical Imaging and Bioengineering (NIBIB) are set to enter Phase II of a multi-phase contract to advance physIQ’s artificial intelligence (AI)-based digital biomarker development for Covid-19.
NCI and NIBIB exercised their option to enter the phase, which brings the total contract to $6.6m.
Being developed to address the fast decompensation of high-risk Covid-19 patients, the tool collects and assesses continuous physiologic data, which can potentially offer early clinical indicators of Covid-19 decompensation.
In Phase I of DeCODe study, physIQ enrolled and analysed 400 high-risk Covid-19 patients in ten weeks.
Using machine learning, the Covid-19 digital biomarker was developed and tested and is based on Phase I data.
In partnership with University of Illinois Hospital and Health Sciences System (UI Health), the DeCODe project will develop an early warning system that enables providers to intervene if the condition of a Covid-19 patient who is clinically surveilled from home is worsening.
Continuous multi-parameter vital signs and physiological features were used to establish a targeted biomarker or Covid-19 Decompensation Index (CDI) for worsening Covid-19.
physIQ chief science officer Stephan Wegerich said: “Using Phase I data, we developed and tested a preliminary digital biomarker, using state of the art machine learning algorithms that take advantage of our extensive library of wearable biosensor analytics as inputs.
“Furthermore, we were able to demonstrate performance levels that far exceeded our target performance criterion. We are looking forward to further validation in Phase II.”
The company noted that Phase II will start this month and enroll 1,200 patients for validating the digital biomarker.
Furthermore, in the validation phase, physIQ will enroll ambulatory Covid-19 patients and analyse the digital biomarker’s ability in predicting decompensation severity remotely.