AI-Powered TB Diagnosis: A Breakthrough in Bringing Healthcare to Rural India

According to a recent report, tuberculosis (TB) cases in India have surged to 2.17 million as of October this year, with the country accounting for approximately 25% of the global TB burden. Improving diagnosis is critical to eradicating TB, and artificial intelligence (AI) can play an crucial role in enhancing detection.

In a bid to tackle this challenge, Hyderabad-based firm Salcit Technologies has developed a product called Swaasa that uses AI to analyze cough sounds and assess lung health. The AI platform is portable and can work on smartphones, tablets, or laptops, providing easy and cost-effective tests to reach larger populations at a higher frequency.

Building on its success, Salcit Technologies has partnered with Google to leverage the Health Acoustic Representations (HeAR) bioacoustic foundation model on Swaasa. HeAR is designed to help researchers build models that can listen to human sounds and flag early signs of disease. Bakul Patel, senior director at Google, stated that this partnership will enable Salcit Technologies to extend screening for TB more widely across India.

Doctors at AIIMS, New Delhi, have acknowledged the need for screening chronic respiratory diseases, citing limited human expertise and lab facilities as a major obstacle. Venkat Yechuri, CEO of Salcit Technologies, claimed that Swaasa has already conducted over 300,000 assessments since its launch, highlighting the potential of AI in improving diagnosis.

The Karnataka Health Promotion Trust (KHPT) has deployed Swaasa in certain districts to enhance screening methods for presumptive TB cases. Additionally, Salcit Technologies is exploring wider deployment in other states, including the North East region.

In a statement, Patel emphasized that "AI is poised to radically transform the way we prevent, diagnose, and treat diseases." This sentiment is echoed by Apollo Radiology International (ARI), which is utilizing image recognition capabilities of Google's AI models for mobile screenings of TB, lung, and breast cancer. The AI systems are also assisting radiologists in interpreting chest X-ray scans for TB.

Google has also licensed its diabetic retinopathy AI model to firms like Forus Health and AuroLab in India, as well as Perceptra in Thailand, aiming to conduct approximately six million AI-assisted screenings in the next 10 years at no cost to patients.