Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
[Preprint]. 2025 Jun 12:rs.3.rs-6630234.
doi: 10.21203/rs.3.rs-6630234/v1.

An Integrated Framework for Automated Image Segmentation and Personalized Wall Stress Estimation of Abdominal Aortic Aneurysms

Merjulah Roby  1 , Juan C Restrepo  1 , Deepak K Shan  2 , Satish C Muluk  2 , Mark K Eskandari  3 , Vikram S Kashyap  4 , Ender A Finol  1
Affiliations

An Integrated Framework for Automated Image Segmentation and Personalized Wall Stress Estimation of Abdominal Aortic Aneurysms

Merjulah Roby et al. Res Sq. .

Abstract

Abdominal Aortic Aneurysm (AAA) remains a significant public health challenge, with an 82.1% increase in related fatalities from 1990 to 2019. In the United States alone, AAA complications resulted in an estimated 13,640 deaths between 2018 and 2021. In clinical practice, computed tomography angiography (CTA) is the primary imaging modality for monitoring and pre-surgical planning of AAA patients. CTA provides high-resolution vascular imaging, enabling detailed assessments of aneurysm morphology and informing critical clinical decisions. However, manual segmentation of CTA images is labor intensive and time consuming, underscoring the need for automated segmentation algorithms, particularly when feature extraction from clinical images can inform treatment decisions. We propose a framework to automatically segment the outer wall of the abdominal aorta from CTA images and estimate AAA wall stress. Our approach employs a patch-based dilated modified U-Net model to accurately delineate the outer wall boundary of AAAs and and Nonlinear Elastic Membrane Analysis (NEMA) to estimate their wall stress. We further integrate Non-Uniform Rational B-Splines (NURBS) to refine the segmentation. During prediction, our deep learning architecture requires 17 ° 0.02 milliseconds per frame to generate the final segmented output. The latter is used to provide critical insight into the biomechanical state of stress of an AAA. This modeling strategy merges advanced deep learning architecture, the precision of NURBS, and the advantages of NEMA to deliver a robust, accurate, and efficient method for computational analysis of AAAs.

PubMed Disclaimer

Conflict of interest statement

Additional Declarations: Competing interest reported. M.R. and the other authors declare that they have no competing interests or personal relationships that could have appeared to in uence the work reported in this manuscript. M.K.E. declares that he has a paid consultancy with W.L. Gore and Boston Scienti c. Competing interest: M.R. and the other authors declare that they have no competing interests or personal relationships that could have appeared to influence the work reported in this manuscript. M.K.E. declares that he has a paid consultancy with W.L. Gore and Boston Scientific.

Figures

Figure 1.
Figure 1.
Framework for automated segmentation and personalized wall stress in AAAs: (a) Architecture of the patch-based dilated U-Net model; (b) Exemplary binary masks overlaid on an AAA image; (c) and (d) Reconstruction and visualization of the AAA in the coronal and sagittal planes; (e) First principal wall stress distribution in a mesh generated using the automated segmentation; stress values are in N/cm2.
Figure 2.
Figure 2.
Volume meshes generated from the ground truth [(a) and (c)] and the predicted [(b) and (d)] image segmentations.
Figure 3.
Figure 3.
Comparative analysis of the maximum hydraulic diameter (in cm) measured using the outer wall boundary.
Figure 4.
Figure 4.
Segmentation of a complex AAA case with metal artifacts: (a) Prediction failure due to metal artifacts (red boxes) affecting automated segmentation and (b) Corrected segmentation using control points of the NURBS curve.
Figure 5.
Figure 5.
SAWS correlation relative to the Ground Truth calculated with Predicted (a) and Predicted + NURBS (b) meshes.
Figure 6.
Figure 6.
Box and whisker plots for the three mesh-based groups comparing five biomechanical parameters.
Figure 7.
Figure 7.
Maximum principal wall stress distribution for an exemplary AAA representative of the Ground Truth (a), Predicted (b), and Predicted + NURBS (c) groups. Stress values are in N/cm2.
Figure 8.
Figure 8.
Comparative analysis of Dice Score Coefficient: existing models vs. present model.
See this image and copyright information in PMC

References

    1. Brown P. M., Zelt D. T. & Sobolev B. The risk of rupture in untreated aneurysms: the impact of size, gender, and expansion rate. J. Vasc. Surg. 37, 280–284, DOI: 10.1067/mva.2003.119 (2003). - DOI - PubMed
    1. Chaikof E. L. et al. The society for vascular surgery practice guidelines on the care of patients with an abdominal aortic aneurysm. J. Vasc. Surg. 67, 2–77.e2, DOI: 10.1016/j.jvs.2017.10.044 (2018). - DOI - PubMed
    1. Lederle F. A. et al. Rupture rate of large abdominal aortic aneurysms in patients refusing or unfit for elective repair. J Am. Med. Assoc. 287, 2968–2972, DOI: 10.1001/jama.287.22.2968 (2002). - DOI - PubMed
    1. Kurugol S., San Jose Estepar R., Ross J. & Washko G. R. Aorta segmentation with a 3d level set approach and quantification of aortic calcifications in non-contrast chest ct. In 2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2343–2346, DOI: 10.1109/EMBC.2012.6346433 (2012). - DOI - PMC - PubMed
    1. Shang E. K. et al. Validation of semiautomated and locally resolved aortic wall thickness measurements from computed tomography. J. Vasc. Surg. 61, 1034–1040, DOI: 10.1016/j.jvs.2013.11.065 (2015). - DOI - PMC - PubMed

Publication types

LinkOut - more resources