Improving the detection and understanding of aortic aneurysms through large-scale radiologic and genetic risk assessments

The National Institutes of Health National Heart Lung and Blood Institute awarded Geisinger Clinic $727,665 on July 15, 2026, under the Cardiovascular Diseases Research program (CFDA 93.837) to improve detection and understanding of aortic aneurysms through large-scale radiologic and genetic risk assessments.

The project uses deep learning-based automated analysis of existing radiologic images to detect and measure aortic aneurysms with high precision, gathering large-scale data on aortic diameters to examine natural history, prevalence, risk factors, and diagnosis rates. The work addresses gaps in current screening guidelines that frequently miss women, non-smokers, and individuals without known family history. In parallel, the award funds integration of genetic risk factors to identify individual genes influencing aortic diameter and uncover genetic variants contributing to abdominal and thoracic aortic aneurysm development and progression. Machine learning and artificial intelligence approaches will identify additional clinical risk factors not previously recognized. Collectively, these efforts will evaluate present screening practices and propose more inclusive diagnostic criteria to reduce underdiagnosis risk.

The project is a Project Grant, an assistance type used for targeted research initiatives within the NHLBI's competitive research portfolio. Performance takes place in Danville, Pennsylvania, with a period of performance from July 15, 2026, through March 31, 2030.

Abstract

Aortic aneurysms are life-threatening enlargements of the aorta that can occur in the abdominal or thoracic segments, sometimes rupturing or dissecting without warning and resulting in up to 80-90% mortality. They often develop silently over many years, making early detection and monitoring essential. While current screening guidelines target many at-risk individuals, they frequently miss important groups such as women, non-smokers, and those without a known family history, leaving them at increased risk. This underdiagnosis leads to increased risk and missed opportunities for early intervention. Our project addresses these gaps by using a deep learning-based automated analysis of existing radiologic images to improve the detection and measurement of aortic aneurysms. By gathering large-scale, high- precision data on aortic diameters, we can examine the natural history, prevalence, risk factors, and diagnosis rates of aortic aneurysms. This work will also evaluate present screening practices and propose more inclusive criteria, aiming to lower the danger of underdiagnosis and potential rupture. In parallel, we will integrate genetic risk factors, building on evidence that certain genetic variants correlate with known aneurysm traits. We will identify individual genes that influence aortic diameter. This gene-focused approach will uncover specific genetic variants that contribute to the development and progression of abdominal and thoracic aortic aneurysms. These data will provide insights into the biological mechanisms driving aneurysm formation and growth at both locations. In parallel, we will employ novel machine learning and AI approaches to uncover additional clinical risk factors that contribute to the development and progression of aortic aneurysms that may have not been previously recognized. These data-driven approaches will leverage the scale of our imaging data, comprehensive electronic health records, and population levels genetic data to better understand how such factors contribute to aneurysm formation and progression in both abdominal and thoracic regions. Combining advanced imaging analysis with genetic insights and expanded clinical risk data, this project will refine screening guidelines, enhance early detection, and broaden our knowledge of the genetic and biological factors and mechanisms underlying aortic aneurysms. Ultimately, by developing more effective strategies for preventing and managing this serious condition, our findings will help save lives and improve patient outcomes.