Collaborative project targets early lung cancer detection in people living with HIV

Lung cancer remains the leading cause of cancer mortality in people living with HIV, the human immunodeficiency virus, with an incidence nearly three times higher than in those without HIV. Vanderbilt engineering professors and physicians from the Vanderbilt University Medical Center are using cutting-edge imaging and biomarker research to advance early detection methods and reduce mortality within the HIV group.

The project, “Evaluation of Radiomic and Blood-Based Biomarkers for the Early Detection of Lung Cancer in People Living with HIV (Biospecimen/Cohort),” seeks to improve lung cancer outcomes by combining radiomic data from low-dose CT scans with blood-based biomarkers. By using machine learning models previously developed at Vanderbilt, the team will assess their predictive power in real-world data from people living with HIV (PLWH). This award is a Cancer Center Support Grant as part of funding for principal investigator Ben Park, M.D., Ph.D., director of the  Vanderbilt-Ingram Cancer Center and Benjamin F. Byrd, Jr. Professor of Oncology.

Yuankai Huo

The new collaboration team includes Yuankai Huo, assistant professor of computer science; Bennett Landman, director of the Vanderbilt Lab for Immersive AI Translation (VALIANT); Lung Screening Program Director Kim Sandler, M.D., professor of radiology and radiological services at the Vanderbilt University Medical Center, Eric Grogan, MD, MPH, associate professor of thoracic surgery, Steve Deppen, PhD, associate professor of thoracic surgery and clinical epidemiologist, and Jessica Castilho, MD, MPH, associate professor of medicine and physician-scientist in infectious diseases at VUMC.

“We’re excited to bring cutting-edge AI technologies into this space, where they can make a real difference for people at high risk of lung cancer,” said Landman, professor of electrical and computer engineering and professor of radiology and radiological sciences.  “This project allows us to leverage our expertise in machine learning and radiomics to solve complex problems in health care.”

“This project will allow us to improve our previous AI models for predicting mortality by adapting it specifically for people living with HIV, using a fully automated analysis of body composition.” said Huo, assistant professor of computer science and electrical and computer engineering.

Bennett Landman

The research project will integrate multi-protein blood panels alongside imaging data to refine diagnostic accuracy. It could potentially revolutionize how early lung cancer detection is approached in this high-risk group, the researchers said. The combination of AI-driven imaging models and blood-based biomarkers holds promise for more personalized, effective screening strategies that could significantly reduce mortality in PLWH.

“With this project, we aim to not only improve lung cancer screening but also ensure that our innovations are relevant and accessible to patients living with HIV,” Sandler said. “We know PLWH are at increased risk for lung cancer, and we have an opportunity to significantly improve care with better early detection.”

Landman said collaboration between engineering and radiology is driving new approaches to health care, showcasing how AI and medical innovation can converge to address urgent public health challenges.

Contact: brenda.ellis@vanderbilt.edu