Early diagnosis is crucial in treating medical conditions. Google has developed an AI model capable of detecting diseases like tuberculosis (TB) and chronic obstructive pulmonary disease (COPD) through cough sounds.

Google introduced the Health Acoustic Representations (HeAR) this year—a bio-acoustic model designed that can analyse sound patterns to produce health insights.

Salcit Technologies, an India-based respiratory healthcare company, has created Swaasa, an AI-powered tool that assesses lung health by analysing cough sounds. They plan to integrate HeAR into Swaasa to enhance its ability to detect TB early.

“Every missed case of tuberculosis is a tragedy; every late diagnosis, a heartbreak. Acoustic biomarkers offer the potential to rewrite this narrative. I am deeply grateful for the role HeAR can play in this transformative journey,” says Sujay Kakarmath, a product manager at Google Research working on HeAR.

The HeAR model, currently available for researchers, has been trained using “300 million pieces of audio data curated from a diverse and de-identified dataset,” and “roughly 100 million cough sounds.”

Despite being curable, TB often goes undiagnosed due to limited access to affordable healthcare. Swaasa, which uses machine learning, could provide a safer and more affordable detection method. Google is also partnering with organisations like the Stop TB Partnership to bring together experts and affected communities with a goal to end TB by 2030.

“Solutions like HeAR will enable AI-powered acoustic analysis to advance tuberculosis screening and detection, offering a potentially low-impact, accessible tool for those who need it most,” says Zhi Zhen Qin, a digital health specialist with the Stop TB Partnership.