An Internet of things device that tracks cough and crowd size in real time could be a useful tool for identifying flu like symptoms in large populations, the researchers said.

The researchers call it flusense, which is about the size of a dictionary. It includes a cheap microphone array, a thermal sensor, a raspberry PI and an Intel movidius 2 neural computing engine.

This idea is inspired by: Using AI technology to analyze audio samples in a given time, determine the number of people in the room and classify them.

Flusense’s system is brilliant at distinguishing cough from other types of non voice audio, so linking cough to the size of a given population can provide useful indicators of how many people are likely to have flu like symptoms.

Can alt4518534235718656 Internet of things technology detect virus

In tests conducted between December 2018 and July 2019, flusense was installed in four waiting rooms at the University of Massachusetts health service clinic, and the researchers said they were able to closely link the results of the system with clinical test symptoms of influenza and other similar diseases.

The lead author of the paper, Dr. forsad al Hossain, and his co-author and consultant, assistant professor tauhidur Rahman, believe that flusense’s bigger plans are in the works.

At present, they are planning to deploy flusense system in many large public places (such as large cafeterias, classrooms, dormitories, gymnasiums, Auditorium) to capture symptom signals from a wide range of people living in a town or city, “they said.

They are also looking for funds to conduct large-scale experiments. At the same time, by extending the function of flusense to capture more symptom signals (for example, sneezing function is added to flusense), which makes the sensing ability more diversified. There is considerable commercial potential in this area of research. “

Flusense is particularly interesting from a technical point of view, as all meaningful processing is done locally via the Intel neural computing engine and raspberry PI. Symptom information is wirelessly sent to the lab for proofreading, but the heavy work is done on the edge.

Al Hossain and Rahman quickly stressed that the device does not collect personally identifiable information – the focus is on aggregating data in a given environment, rather than identifying any patient’s personal privacy data, and that the information it collects is highly encrypted, ensuring privacy.

According to the researchers, flusense focuses on treating it as a health monitoring tool rather than a diagnostic device.

Al Hossain and Rahman say it has several important advantages over other health monitoring technologies, especially Internet based tracking technologies such as Google flu trend and twitter.

“Flusense is less susceptible to public health campaigns or advertising. Moreover, the contactless nature of the sensor is well suited to capture symptom signals from different geographic locations and groups, including vulnerable groups who may not have access to health care and may not be able to see a doctor / clinic. “

Editor in charge: CT

Leave a Reply

Your email address will not be published. Required fields are marked *