MIT AI Detects Asymptomatic COVID-19 by Analyzing Cough Sounds
MIT researchers develop an AI model that identifies COVID-19 in asymptomatic individuals with perfect accuracy by analyzing the subtle acoustic patterns of their coughs.
MIT researchers develop an artificial intelligence model that identifies COVID-19 in asymptomatic individuals by analyzing the subtle acoustic patterns of their coughs. The team builds what they believe is the largest research cough dataset to train this AI, achieving a remarkable 100% detection rate for asymptomatic carriers and 98.5% for symptomatic cases.
The AI system detects minute changes in vocal strength, lung performance, and muscular degradation that are invisible to the human ear. This approach builds on previous medical research where AI models successfully identify conditions like pneumonia, asthma, and even Alzheimer's disease solely through cough sounds.
Despite the impressive accuracy, the lead researcher cautions that this tool functions as a screening method rather than a formal diagnostic tool for symptomatic patients. The technology focuses specifically on distinguishing COVID-19 from other conditions, offering a promising early warning system to help curb silent transmission of the virus.