Would you want to know if you have tuberculosis without going to the doctor or having your lungs examined? Well, Google AI can do this for you.
Google released research on the Cornell University website in March 2024 that covered their HeAR artificial intelligence. The basic model was trained on 300 million audio recordings, according to the business. Specifically, they trained the cough model with around 100 million cough noises. As a result, HeAR can recognize patterns in sounds associated with health for medical audio analysis.
The Health Acoustic Representations model works well for a variety of applications and microphone types. Consequently, it has an enhanced capacity to discern significant patterns in audio data associated with health.
Google provided academics with access to its AI for health analysis so they could create comparable products more quickly. Salcit Technologies, which created the Swaasa AI software, was one of these researchers. To help with diagnosis, it employs artificial intelligence to identify noises associated with TB. Because Swaasa lowers expenses and eliminates the need for specialist equipment, more individuals can obtain lung treatment.
In a statement, Sujay Kakarmarth, a product manager at Google Research working on HeAR said, “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.”
According to Google, this technology is supported by The StopTB Partnership, a UN-hosted organization that aims to eradicate tuberculosis by 2030. Zhi Zhen Qin, its digital health specialist, posted the following: “Solutions like HeAR will enable AI-powered acoustic analysis to break new ground in tuberculosis screening and detection, offering a potentially low-impact, accessible tool to those who need it most.”