What’s in your cough? An assessment of your lung health, generated by AI in smartphone app Swaasa

At JIPMER Puducherry, researchers have used Swaasa specifically for the detection of pulmonary tuberculosis. AIIMS Ballabgarh is currently piloting the tool in Uttarakhand to screen individuals at risk for COPD.

Swaasa, which translates as Breath, combines the analysis of sounds with data on symptoms and demographics to provide a preliminary diagnosis. It is on the list of the 10 high-impact AI tools developed in India in the official compendium of the ongoing AI Impact Summit 2026. It has the potential to significantly improve screening and detection for lung disease and to bridge healthcare gaps especially in rural and remote areas.

“The recorded cough is analysed using an AI algorithm, alongside symptom-based history and basic demographic details. Based on this combined information, the system identifies whether the person is likely to have obstructive or restrictive lung disease, including conditions such as COPD and asthma,” Dr Harshal Ramesh Salve, additional professor at the AIIMS Centre for Community Medicine, told The Indian Express.

Chronic cough, Dr Salve said, carries characteristic sound patterns. “The AI algorithm recognises acoustic signatures from the sound waves, and can differentiate between types of lung disease by analysing these patterns,” he said.

Swaasa, Dr Salve said, has undergone extensive validation in primary and secondary care settings over the last 3-4 years.

“We have tested it in hospitals that provide multiple levels of care and also evaluated it at AIIMS. The validation at AIIMS is particularly significant,” he said. Swaasa has also been deployed at the state level in several states such as Karnataka, Dr Salve said.

Narayana Rao Sripada, the founder of Salcit Technologies, the company behind Swaasa, said the tool was conceptualised in 2017 in recognition of the challenges of screening for respiratory diseases in community settings.

“In many rural and remote areas, spirometers (devices that measure lung function by monitoring air volume and flow rate during breathing) are either unavailable or require trained personnel. There is no systematic way to identify people at risk of developing respiratory illness. We explored whether cough acoustics could serve as a simple, non-invasive screening tool,” Sripada said.

Coughs, he said, are physiological responses triggered by receptors in the respiratory system. “Depending on where these receptors are activated — upper airways in asthma, deeper in the lungs in COPD — the cough exhibits distinct acoustic characteristics. Variations in glottis and respiratory structures across diseases are reflected in sound waves,” he said.

The development of Swaasa involved collecting a large database of cough sounds from both healthy individuals and patients with respiratory diseases, and training AI models to recognise disease-specific patterns. Data were collected at institutions such as Nizam’s Institute of Medical Sciences and Apollo Hospitals, Hyderabad.

The development of the prototype was supported by grants from the Department of Biotechnology in the Ministry of Science and Technology, and the Ministry of Electronics and Information Technology, and from international agencies including the United Kingdom’s Foreign, Commonwealth and Development Office (FCDO).

Sripada underlined that Swaasa is not intended to replace spirometry, but to only serve as a screening and diagnostic aid in peripheral health centres.

“The software first determines whether a person’s respiratory system is normal or abnormal. If abnormal, it classifies the condition as obstructive or restrictive and further narrows down to the likelihood of specific diseases such as asthma, COPD, or pulmonary tuberculosis,” he said.

According to Sripada, the app has demonstrated an overall accuracy of about 90% in identifying respiratory abnormalities. It can differentiate between obstructive and restrictive conditions with roughly 85% accuracy, has shown around 83% accuracy in detecting asthma or COPD, and about 79% with pulmonary tuberculosis.

“Several NGOs in Bihar are using the platform for rural healthcare screening,” Sripada said. “AIIMS Ballabgarh is using the tool in Uttarakhand in collaboration with the [philanthropic trust] ALVL Foundation and the [pharma giant] Cipla, which supports the initiative with spirometry tools for comparative testing,” he said.

The product, which is patented and endorsed by reputed medical journals, is officially classified as a regulated software medical device, Sripada said.