Follow Datanami:
January 15, 2021

AI-Powered, Wearable-Enabled COVID-19 Screening Solution Begins Validation

Despite millions of tests per day, the United States remains mired in the darkest phase of the COVID-19 pandemic, with thousands of daily deaths and rising case numbers. One heavy-hitter contributor to this crisis is COVID-19’s propensity for asymptomatic transmission, which medical researchers estimate may account for around 59% of all COVID-19 transmission. Of course, most without symptoms won’t choose to wait in line and get tested – so how do we detect those asymptomatic cases early enough to stop further transmission and better treat presymptomatic patients? AI firm NeuTigers has teamed up with networking firm Rajant Corporation to deliver an AI- and wearable-powered solution called CovidDeep, which is now in validation following a “promising” clinical trial.

“Right now, our biggest question is, how can we return to an open society and a productive economy without risking the collapse of the healthcare system? The answer is through a scalable SARS-CoV-2/COVID-19 screening system, which we knew we could bring to life with Rajant as our partner,” says Adel Laoui, CEO and founder of NeuTigers

The Empatica E4 wristband. Image courtesy of Empatica.

In a case study, NeuTigers explained how the system works. In partnership with the San Matteo Hospital in Pavia, Italy, the company collected digital physiological data from a Samsung Galaxy S8 smartphone, a pulse oximeter and a paired Empatica E4 smartwatch – an off-the-shelf device that measures a variety of physiological signals such as blood volume pulse, skin temperature, and the electrical properties of the wearer’s skin. 

The collected data (supplemented by a questionnaire) is then denoised and preprocessed. Using that prepared data, CovidDeep then applies a proprietary AI algorithm to detect different types of COVID-19 infection: healthy, asymptomatic, or symptomatic.

Based on the controlled clinical trial, which involved 87 participants, NeuTigers concluded: “Our system achieves 98% accuracy in distinguishing among healthy, versus asymptomatic, and symptomatic COVID-19 subjects. We have demonstrated that CovidDeep can be employed in a pervasive fashion across a diverse set of mobiles and wearable devices while offering high efficiency and accuracy.”

“CovidDeep does not require the presence of a nurse or physician during testing,” said Ignazio R. Marino, executive vice president of Jefferson International Innovative Strategic Ventures and co-author of the study. “It uses physiological signals extracted using commercially available devices and achieves high test accuracy. As a result, it provides a testing mechanism that is accurate, easily accessible to the public, and easy for individuals to use. Furthermore, this mechanism only requires a few minutes of data collection from an individual to perform an inference. … CovidDeep has the potential to impact how we screen for SARS-CoV-2, and subsequently triage the use of other diagnostic resources.”   

Now, CovidDeep is moving into the validation phase with a second study. NeuTigers is now seeking volunteers for the study, which it says is “especially critical to frontline healthcare workers.”