A new artificial intelligence app can detect serious respiratory diseases by analyzing the unique acoustic signature of your cough, potentially revolutionizing how millions of people worldwide get screened for conditions like tuberculosis, asthma, and COPD (chronic obstructive pulmonary disease). The app, called Swaasa, costs as little as $1.50 per test and works on any smartphone, making it accessible in areas where traditional lung testing equipment is unavailable. Why Millions Skip Lung Disease Screening? In India, which accounts for roughly one quarter of the world's tuberculosis burden, the screening gap is staggering. The National TB Prevalence Survey found that 63.6% of people with tuberculosis symptoms never sought medical help. The reason is often heartbreakingly simple: traditional screening requires traveling to a hospital or lab, taking time off work, and losing a day's wages. For people living in rural areas or working in informal jobs, this barrier is insurmountable. The numbers reflect this reality. While India's government aims to screen at least 90% of people at higher risk for tuberculosis, research by the Indian Council of Medical Research found that fewer than 8% are actually screened. This gap means preventable and curable diseases go undetected, allowing tuberculosis to kill more than a million people globally each year despite decades of public health efforts. How Does a Cough Sound Reveal Lung Disease? Every cough has a distinct acoustic pattern, much like a fingerprint. Research shows that different respiratory conditions produce recognizable cough signatures. For example, multiple forceful coughs can indicate asthma or bronchitis, while multiple weak coughs may suggest pneumonia. Other conditions like widened or narrowed airways, inflamed bronchial tubes, fluid in air sacs, and lung scarring each produce their own characteristic sound patterns. Narayana Rao Sripada, a technology expert who co-founded Swaasa, spent months listening to people cough in hospital settings in Hyderabad, India. He identified the acoustic markers associated with various respiratory conditions and trained a machine learning model to recognize them. When users download the app and cough three to four times into their phone, the artificial intelligence analyzes the sound and assesses their risk within 30 seconds. The accuracy is impressive. A 2025 study comparing Swaasa's diagnoses with those of a pulmonary physician found that the app correctly identified respiratory issues in 87.32% of cases. The app can detect respiratory problems with around 89 to 90% accuracy overall, and importantly, it produces very few false negatives, meaning it rarely misses actual cases of disease. How to Use Swaasa for Respiratory Screening - Download the app: Swaasa is available on smartphones and requires no special equipment beyond the phone's built-in microphone. - Cough into your phone: Users cough three to four times into the app, which captures the acoustic data of their cough sounds. - Get instant results: Within 30 seconds, the app analyzes the cough and provides a risk assessment for respiratory diseases including tuberculosis, asthma, COPD, bronchitis, and pneumonia. - Follow up if needed: If the app flags a higher risk, users can take the results to a healthcare provider for further testing and diagnosis. The app is designed to work in real-world conditions where traditional testing isn't available. It can make at least 360 assessments on a single phone charge and works in areas with poor network connectivity. The developers estimate it could scale to perform one million assessments daily and potentially reach 500 million people. Real-World Impact in Mobile Health Clinics In Behror, a hot, dusty town in Rajasthan, India, the nonprofit Child Survival India operates mobile medical vans that bring healthcare to communities. When patients arrive with symptoms like persistent cough or unexplained weight loss, or when they've been in contact with someone who has tuberculosis, the mobile health workers now use Swaasa to screen them. "For many who come to our mobile medical units, this means taking time off from work, or forgoing a day's wages. Many just won't do it," said Deepa Bajaj, president of Child Survival India. Deepa Bajaj, President, Child Survival India The results have been encouraging. Child Survival India medics have used the app to screen over 800 patients and find it useful for motivating people to seek further care. When patients see on their phone that the app suggests they're at higher risk, they're more likely to visit a chest clinic for confirmation. This simple visual feedback has proven more persuasive than traditional health worker recommendations. What Makes Swaasa Different From Other AI Health Tools? Artificial intelligence is increasingly used in healthcare to screen for diseases. Around the world, AI systems analyze lung X-rays and ultrasound images to detect tuberculosis. Similar technology screens for diabetic eye disease and parasitic infections in rural Kenya. But Swaasa stands apart because it analyzes actual cough sounds rather than images or text data. This approach offers practical advantages. Traditional spirometry, a common lung function test, requires specialized equipment and trained professionals to operate. Many low-resource settings in India and across the Global South lack these resources. Swaasa requires only a smartphone, making it far more scalable and accessible. The cost difference is dramatic. A single Swaasa assessment costs as little as $1.50, with special rates available for community health providers. Compare this to the expense and logistical burden of traditional screening, and the advantage becomes clear. "The unique selling proposition of Swaasa is that you can scale it, and scale it very fast. You can screen a large number of people in a very short span of time with a reasonable sensitivity," said Professor Rakesh Kumar, who ran early tests on the app at the Centre for Community Medicine in Delhi's All India Institute of Medical Sciences. Professor Rakesh Kumar, Centre for Community Medicine, All India Institute of Medical Sciences Professor Kumar noted another important benefit: the app produces very few false negative results, meaning it rarely misses cases of respiratory disease. By flagging only those individuals who need further testing, Swaasa reduces the number of X-rays and other expensive tests needed, saving both time and resources. The app is particularly effective at identifying asymptomatic patients who have no obvious symptoms but carry disease. Where Is Swaasa Being Used Now? Over 400,000 individuals have been tested using the Swaasa app so far, with monthly usage exceeding 5,000 tests. Pilot screening programs for tuberculosis are underway in several Indian states, including Andhra Pradesh in southern India and Bihar in central India. The company behind Swaasa, Salcit Technologies, is collaborating with India's first AI-enabled government hospital and incubation center in Uttar Pradesh to launch large-scale respiratory health screening initiatives in 2026. The nonprofit Healing Fields is training community health entrepreneurs in villages across Bihar, Uttar Pradesh, and Jharkhand to use Swaasa. These health workers provide accessible, affordable primary care by connecting patients to qualified doctors through telehealth consultations, and the app extends their screening capabilities. The app has also been tested for detecting COVID-19, suggesting its potential extends beyond tuberculosis and chronic lung diseases. An Australian company called Helfie is developing health AI to screen for over 20 conditions and has licensed Swaasa specifically to screen for respiratory disorders. What Challenges Remain? Despite its promise, Swaasa faces barriers to widespread adoption. The app is designed for use and interpretation by healthcare providers, but convincing doctors to embrace a new diagnostic tool has proven difficult. Cough analysis is a new marker and testing method that physicians haven't learned in their regular medical training. Changing medical practice requires not just a better tool, but also education and cultural shift within the healthcare system. The app has been recognized as a Class B medical device in India, but it still awaits approval from the U.S. Food and Drug Administration. This regulatory step will be important for expanding its use in the United States and other countries with strict medical device oversight. Despite these hurdles, the founders remain optimistic. They're collaborating with Google's HeAR AI model, which has been trained on millions of audio samples, to further improve the app's accuracy over time. As more people use Swaasa, the machine learning model will continue to learn and refine its ability to detect respiratory disease from cough sounds.
