Asymptomatic transmission of COVID-19 has been one of the major problems. However, now, a newly developed artificial intelligence (AI) model can detect asymptomatic COVID-19 cases just from the sound of people’s cough.
Researchers at MIT developed an AI model that can detect asymptomatic COVID-19 cases just by listening to subtle differences in your coughs. They will now test the model in human trials after seeking approval from the U.S. Food and Drug Administration (FDA), which could be used as an effective screening tool.
The AI model is based on previous models the researchers developed to detect pneumonia, asthma, and Alzheimer’s disease. It is actually the Alzheimer’s AI model that the researchers used in an effort to detect COVID-19.
Study co-author Brian Subirana said, “The sounds of talking and coughing are both influenced by the vocal cords and surrounding organs. Things we easily derive from fluent speech, AI can pick up simply from coughs, including things like the person’s gender, mother tongue or even emotional state. There’s in fact sentiment embedded in how you cough.”
The researchers found that their AI model was able to pick up subtle differences in the coughs associated with four characteristic features of COVID-19 – “muscular degradation, vocal cord strength, sentiment such as doubt and frustration, and respiratory and lung performance.”
The team’s AI model accurately identified more than 98% of people with COVID-19 infection and correctly ruled out COVID-19 in over 94% of people without the infection.
Interestingly, for asymptomatic people, the AI model correctly identified 100% of people with COVID-19 and ruled out COVID-19 in more than 83% of people without the infection, according to Live Science.
The findings of the study were published on September 30 in the IEEE Open Journal of Engineering in Medicine and Biology.
Dr. Anthony Lubinsky of NYU Langone Tisch Hospital, who was not involved in the study, said these are “a pretty encouraging set of numbers,” and the results are “very interesting.”
However, he told Live Science, “whether or not this performs well enough in a real-world setting to recommend its use as a screening tool would need further study.”
The MIT team is not the only one working on detecting COVID-19 through cough sounds. Researchers from Cambridge University, Carnegie Mellon University, and the U.K. start-up Novoic have also been using similar techniques to detect COVID-19, according to BBC.
The study authors wrote, “Pandemics could be a thing of the past if pre-screening tools are always-on in the background and constantly improved. Those always-listening tools could be smart speakers or smartphones.”
The article originally appeared on Live Science.
