Wireless Sensing Technology Enables Touch-Free Diagnostics of Common Lung Diseases

Diagnosing lung diseases often requires physical contact, imaging scans, or specialized equipment, which can limit access in low-resource settings and increase infection risk during outbreaks. Subtle changes in breathing patterns can signal conditions such as asthma, pneumonia, or tuberculosis, but these are difficult to capture reliably without direct examination. Researchers have now demonstrated a contactless diagnostic approach that can accurately identify multiple pulmonary diseases by analyzing breathing patterns using wireless signals and artificial intelligence (AI).

The system, developed by an international team of engineers and computing scientists from the Information Technology University (Lahore, Pakistan) the University of Glasgow (Scotland, UK), uses harmless microwave radio signals transmitted at 5.23 GHz, a frequency aligned with future 6G and Wi-Fi 7 networks, to sense chest movements linked to breathing. Reflected signals are analyzed using advanced AI models to extract disease-specific respiratory signatures without touching the patient.

Patients are exposed to radio waves emitted by software-defined radios, and the system captures how those signals are altered by breathing motion. Multiple machine learning and deep learning models interpret these reflections to distinguish between normal and abnormal respiratory patterns. The approach leverages integrated sensing and communication, allowing the same wireless infrastructure to transmit data while performing health sensing.

The researchers evaluated the system using microwave reflection data from 190 patients with diagnosed respiratory diseases and 30 healthy individuals as controls. Data were collected across two periods, including a high-smog season, resulting in nearly seven and a half hours of real-world radio data. In laboratory testing, the system screened for asthma, chronic obstructive pulmonary disease, interstitial lung disease, pneumonia, and tuberculosis with 98% accuracy, with a deep learning vanilla CNN model performing best.

This wireless sensing approach, presented in Communications Medicine, could enable continuous, contactless respiratory monitoring in hospitals, homes, and smart living environments. Because it uses only a fraction of available wireless bandwidth, it could be integrated into future 6G networks without disrupting data transmission. The technology may be especially valuable in low-resource settings and during infectious disease outbreaks, where reducing physical contact can limit disease spread. Future work will focus on broader clinical validation and deployment in real-world environments.

“This research showcases the effectiveness of ISAC, which allows a single communications infrastructure to both transmit data and perform sensing tasks at the same time,” said Professor Qammer H. Abbasi, who led the research and is one of the paper’s corresponding authors. “The sophisticated sensing which underpins our results only took up 12.5% of the system’s available bandwidth. That means that the rest of the system’s bandwidth could be used for data transmission to help enable future generations of integrated, continuous health monitoring devices.”

Related Links:
University of Glasgow