India-based respiratory healthcare company Salcit Technologies is collaborating with the Google Research team to explore how Google's Health Acoustic Representations (HeAR) can help extend the capabilities of Salcit's bioacoustic AI technology, Swaasa.
Swaasa is using HeAR to help enhance research and early detection of TB based on cough sounds.
“Missing TB is a tragedy, and a delayed diagnosis is heartbreaking,” Sujay Kakarmath, a product manager at Google Research working on HeAR, said in a statement. “Acoustic biomarkers have the potential to rewrite this narrative, and we're deeply grateful for the role HeAR can play in this transformative journey.”
The Google team trained HeAR using 300 million pieces of audio data curated from a diverse, anonymized dataset, and specifically trained the cough model using around 100 million cough sounds.
The company says HeAR will learn to identify health-related sound patterns, building a powerful foundation for medical speech analytics.
Salcito said Swatha has a history of using machine learning to help detect disease earlier, while also bridging the gap in accessibility, affordability and scalability by providing location- and device-independent respiratory health assessments.
Leveraging HeAR, Salcit aims to scale up TB testing more widely across India.
Additionally, Google Research said it has received support for its AI-driven TB efforts from a United Nations organization called the End TB Partnership, which brings together TB experts and affected communities with the goal of eliminating the disease by 2030.
“Solutions like HeAR, enabling AI-powered acoustic analysis, will break new ground in TB screening and detection, providing low-impact, easily accessible tools to those who need them most,” Qin Zhizhen, digital health expert at the Stop TB Partnership, said in a statement.
Larger trends
Over the past few years, the use of voice and machine learning technologies to diagnose and monitor a variety of health conditions has become increasingly popular.
In 2022, smart stethoscope manufacturer EKO received FDA clearance for its algorithm to detect and characterize heart murmurs in adult and pediatric patients. Eko Murmur Analysis Software is a machine learning algorithm that uses heart sounds, phonocardiograms, and ECG signals to detect harmless and structural heart murmurs.
Israeli medical technology company TytoCare has raised $49 million to fund the growth of its AI-powered TytoCare Home Smart Clinic, which allows doctors to perform remote consultations and includes connected devices that collect readings from an otoscope, tongue depressor, thermometer and an FDA-approved stethoscope that analyzes lung sounds to detect wheezing.
Canary Speech, a speech analytics software developer, has partnered with Microsoft to apply AI techniques to extend machine learning speech models for healthcare. Canary offers voice biomarker technology that collects and analyzes data to determine whether an individual's voice has abnormalities.
Under the terms of the partnership, Canary will employ Microsoft AI to accelerate its speech analytics technology to reduce healthcare costs, address mental health challenges and scale remote patient monitoring solutions.
The HIMSS AI in Healthcare Forum is scheduled to take place in Boston on September 5-6. More information and registration available here.
