Researchers develop a device able to monitor trends in flu-like illnesses to aid prevention mechanisms

Scientists from the University of Massachusetts developed a portable device, FluSense, powered by artificial intelligence to detect coughing and crowd size in real time. The device analyses the data which are used to monitor trends in flu-like illnesses such as Covid19.

The FluSense platform processes a low-cost microphone array and thermal imaging data. Personally identifiable information, such as speech data or distinguishing images, are not stored on the platform. The researchers developed a lab-based cough model and then trained the algorithm to create thermal images representing people, and then count them.

The device is meant for hospitals, healthcare waiting rooms and larger public spaces. The researchers noted that models like this one can be lifesavers by informing the public health response during a flu epidemic. Data collected can help in determining the timing for flu vaccine campaigns, potential travel restrictions, the allocation of medical supplies and more.

FluSense was placed in a rectangular box about size of a large dictionary, in four healthcare waiting rooms. From December 2018 to July 2019, it collected and analysed more than 350,000 thermal images and 21 million non-speech audio samples from public waiting areas.

"This may allow us to predict flu trends in a much more accurate manner," said Tauhidur Rahman, co-author of the study published in the journal Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies.

The researchers found that FluSense was able to predict daily illness rates at the university clinic. "I thought if we could capture coughing or sneezing sounds from public spaces where a lot of people naturally congregate, we could utilise this information as a new source of data for predicting epidemiologic trends," Rahman added.

Adapted from: https://yourstory.com/2020/03/portable-ai-device-monitor-coronavirus-cov...
For more information: https://dl.acm.org/doi/10.1145/3381014