Google developed an AI technology called Health Acoustic Representations (HeAR) that can detect early signs of respiratory diseases like tuberculosis and COPD by analyzing subtle changes in coughs, sniffles, breathing, and other bioacoustic signals, even using simple microphones like those in smartphones.
The AI model is trained on millions of audio samples and can distinguish acoustic signals with adaptability and versatility, making it a promising tool for preventive medicine and equitable healthcare access.
The development of bioacoustics-based AI models like HeAR could have an impact on global health by reducing reliance on expensive and hard-to-access equipment, promoting early disease detection and monitoring, and supporting the efforts of WHO and other organizations to eradicate tuberculosis.
Google is also collaborating with Chang Gung Memorial Hospital in Taiwan to develop ultrasound-based AI models for early detection of breast cancer, aiming to provide free screenings to underserved populations.
This article was written in collaboration with Generative AI news company Alchemiq
Sources: Mashable, La Capital, Gigazine, Insomnia.gr, Heise