Google's AI model HeAR, in partnership with Salcit Technologies, detects early tuberculosis signs with 94% accuracy using audio samples.

Google's AI model, Health Acoustic Representations (HeAR), identifies respiratory illness sounds, such as coughs, sniffles, and labored breathing, through 300 million audio sample training. Partnering with Indian health tech company Salcit Technologies, HeAR's data is used in Salcit's app, Swaasa, to detect early signs of tuberculosis with 94% accuracy. This low-cost, non-invasive technology could revolutionize healthcare access in regions with limited resources.

August 30, 2024
333 Articles

Further Reading