AI-Powered Biosensor Technology to Enable Breath Test for Lung Cancer Detection

Detecting lung cancer early remains one of the biggest challenges in oncology, largely because current tools are invasive, expensive, or unable to identify the disease in its earliest phases. Researchers are now working on a noninvasive solution: a breath-based screening system that analyzes volatile organic compounds (VOCs) linked to thoracic cancers.

The technology, developed at the University of Texas at Dallas (Richardson, TX, USA) in collaboration with UT Southwestern Medical Center (Dallas, TX, USA), uses a biosensor paired with artificial intelligence (AI) to detect chemical signatures associated with lung and esophageal cancers. The system, built by experts in bioengineering, computer science, and clinical pulmonology, combines an electrochemical biosensor with machine learning analysis.

Image: The screen-printed electrochemical sensor (Photo courtesy of UT Dallas)

The biosensor identifies eight VOCs that may serve as biomarkers for thoracic malignancies, while AI evaluates the biochemical characteristics of these compounds to determine whether they match known cancer profiles. The breath-based platform leverages changes in metabolites that appear early in disease onset, making it suitable for screening. Inspired partly by the search for rapid, noninvasive COVID-19 detection methods, the device uses breath sampling because exhaled air carries metabolites that signal underlying health conditions. To test the device, researchers analyzed breath samples from 67 individuals, including 30 with biopsy-confirmed thoracic cancer.

The findings, published in Sensing and Bio-Sensing Research, show that the biosensor accurately detected the relevant VOCs in 90% of confirmed cases. AI played a critical role in interpreting complex data from breath samples, with machine learning models refined and validated through collaboration between engineering and computer science teams. The device also benefited from clinical insight from pulmonology specialists, ensuring that real-world disease patterns guided its development.

If validated in larger patient cohorts, the technology could serve as a fast, affordable, and noninvasive screening tool for primary care settings. Its ability to help identify cancer risk earlier could support timely referrals and reduce the burden of invasive diagnostic procedures. The system may ultimately complement annual health examinations and routine bloodwork, offering clinicians an additional method to flag potential thoracic malignancies.

“We built a screening tool that could allow physicians to catch the disease in its early phases, which improves outcomes,” said Dr. Shalini Prasad, corresponding author of the study. “This technology offers a potentially affordable, quick, and noninvasive breath analysis tool for cancer screening.”

Related Links:
UT Dallas
UT Southwestern Medical Center