Advertisement

Scientists develop AI device that detects coughs in crowds

Tauhidur Rahman, left, and Forsad Al Hossain display the FluSense device they invented. Photo courtesy of University of Massachusetts Amherst
Tauhidur Rahman, left, and Forsad Al Hossain display the FluSense device they invented. Photo courtesy of University of Massachusetts Amherst

March 19 (UPI) -- University of Massachusetts Amherst scientists unveiled Thursday a new artificial intelligence device, FluSense, they invented to detect coughing patterns in crowds, which can then be analyzed to forecast flu-like trends.

Inventors have envisioned the device being used in hospitals, waiting rooms and larger public spaces.

Advertisement

They say it may expand on health surveillance tools already used to forecast seasonal flu and other viral respiratory outbreaks, like the coronavirus pandemic or SARS outbreaks. And FluSense may also inform public health response with regard to potential travel restrictions and allocation of medical supplies.

"This may allow us to predict flu trends in a much more accurate manner," said inventors Tauhidur Rahman, assistant professor of computer and information sciences, and doctoral student Forsad Al Hossain.

Rahman is also Hossain's adviser and the co-author of the FluSense study published Wednesday in the Association for Computing Machinery on Interactive. Mobile, Wearable and Ubiquitous Technologies, Hossain is the lead author.

The device uses a microphone array, thermal camera and neural computing engine to capture "cough sounds" and crowd size changes in "real-time," the study's abstract say.

"We developed a contactless syndromic surveillance platform FluSense that aims to expand the current paradigm of influenza-like illness surveillance by capturing crowd-level bio-clinical signals directly related to physical symptoms of ILI from hospital waiting areas in an unobstrusive and privacy-sensitive manner," the abstract says. "FluSense consists of a novel edge-computing sensor system, models and data processing pipelines to track crowd behaviors and influenza-related indictors, such as coughs, and to predict daily ILI and laboratory-confirmed influenza caseloads."

Advertisement

The device was tested over seven months in four public waiting areas within the university hospital. The testing, from December to July 2019, analyzed more than 350,000 thermal images and 21 million audio samples from the waiting rooms and found that it was accurately able to predict illness rates at the university clinic.

The FluSense inventors partnered with Dr. George Corey, executive director of University Health Services; bio-statistician Nicholas Reich, director of UMass based CDC Influenza Forecasting Center of Excellence; and epidemiologist Andrew Lover, a vector-borne disease expert and assistant public health professor, to try out the device.

Testing out FluSense in other public areas and geographic locations is the next step.

"We have the initial validation that the coughing indeed has a correlation with influenza-related illness," Lover said. "Now we want to validate it beyond this specific hospital setting and show that we can generalize across locations."

Latest Headlines