Northwestern creates wearable sensor for detecting COVID-19 symptoms

Northwestern researchers have developed a throat sensor to measure vibrations from coughs and chest movements to track early signs of COVID-19 via AI. (Northwestern University)

Engineering researchers at Northwestern University are developing a small sensor that sticks to a person’s throat and detects cough patterns, respiratory sounds, heart rate and body temperature.

The postage stamp-sized sensor transmits continuous data streams via Bluetooth to a tablet or smartphone and then to the cloud for analysis via artificial intelligence (AI). Researchers hope to find patterns that indicate COVID-19 infections, potentially even before a person does, and eventually to help doctors develop treatment protocols.

Northwestern Prof. John Rogers led the technology’s development. The use of a sensor on the suprasternal notch on a person’s throat offers more accurate readings than a wearable watch or ring on the wrist or finger.

Sponsored by Infosys

In Conversation with Antonio Neri, President & CEO – Hewlett Packard Enterprise & Salil Parekh, CEO – Infosys

Hear the CEOs of Infosys & HPE discuss the current crisis and how it has accelerated the need for digital transformation for their clients. Connectivity at the Edge, right mix of hybrid cloud, ability to extract data faster than ever before… these are just some of the contributions that HPE and Infosys make to our clients’ digital transformation journey.

“There are many ideas for wearables to track the disease, but none addresses the key symptoms of fever, cough and respiratory anomalies,” he said via email to FierceElectronics. “These features cannot be measured from the wrist via Fitbit or Apple Watch or the finger via the Oura ring.”

The sensor checks vibrations rather than acoustics and was originally developed for monitoring swallowing and speech disorders in patients recovering from stroke.  It monitors vibrations like chest wall movements and cough, which Northwestern hopes to refine with machine learning. Rogers said his team refers to the device as a "mechano-acoustic sensor" which measures body sounds and not acoustic waves in the air.

A newer design shrinks the original postage stamp size down to 2 x 2 mm to sit on the suprasternal notch. From there, the sensor connects to the radio and memory unit on the chest via a narrow umbilical. This makes the resulting upper device nearly invisible, Rogers said.

The sensors have the potential to offer frontline medical workers and patients information to reduce the risks of transmission, said Prof. Arun Jayaraman, a scientist at Northwestern’s Shirley Ryan AbilityLab.

“People with obvious, severe symptoms going to the hospital, being tested or being told to self-isolate,” Jayaraman said in a statement. “For those who have symptoms they perceive as mild or seasonable allergies, there is no warning system. They could be in contact with others and unknowingly spread infection.”

“We feel there are many opportunities for AI…and we’re currently only scratching the surface,” Rogers added.

The throat sensor was first tested on 25 people in April and is now being tested on 50 people, some who have been treated at Northwestern Memorial Hospital in Chicago and have been sent home, Rogers said on Thursday. While the device is still in early stages of development, several large companies are exploring deployments at a bigger scale.

The research team launched a small engineering company called Sonica Health based on the intellectual property involved.  Uses of the device in COVID-19 response are supported by the Biomedical Advanced Research and Development Authority (BARDA) inside the U.S. Department of Health and Human Services.

RELATED: 

Wanted: an audio recording of your cough

Monitoring newborn babies with wireless sensors

 

Suggested Articles

The world’s largest chipmaker saw a 47% decline in data center sales to enterprise and government, even as it forecast a full year 2020 record of $75B

Working with Jacoti of Belgium, Qualcomm wants to make earbuds recognize the hearing anomalies of users.

Deep learning is one of the most promising techniques for training machines to "think" like people.