Automated Aneurysm Boundary Detection and Volume Estimation Using Deep Learning

<b>Background/Objective:</b> Precise aneurysm volume measurement offers a transformative edge for risk assessment and treatment planning in clinical settings. Currently, clinical assessments rely heavily on manual review of medical imaging, a process that is time-consuming and prone to i...

Full description

Saved in:
Bibliographic Details
Main Authors: Alireza Bagheri Rajeoni, Breanna Pederson, Susan M. Lessner, Homayoun Valafar
Format: Article
Language:English
Published: MDPI AG 2025-07-01
Series:Diagnostics
Subjects:
Online Access:https://www.mdpi.com/2075-4418/15/14/1804
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849521430069248000
author Alireza Bagheri Rajeoni
Breanna Pederson
Susan M. Lessner
Homayoun Valafar
author_facet Alireza Bagheri Rajeoni
Breanna Pederson
Susan M. Lessner
Homayoun Valafar
author_sort Alireza Bagheri Rajeoni
collection DOAJ
description <b>Background/Objective:</b> Precise aneurysm volume measurement offers a transformative edge for risk assessment and treatment planning in clinical settings. Currently, clinical assessments rely heavily on manual review of medical imaging, a process that is time-consuming and prone to inter-observer variability. The widely accepted standard of care primarily focuses on measuring aneurysm diameter at its widest point, providing a limited perspective on aneurysm morphology and lacking efficient methods to measure aneurysm volumes. Yet, volume measurement can offer deeper insight into aneurysm progression and severity. In this study, we propose an automated approach that leverages the strengths of pre-trained neural networks and expert systems to delineate aneurysm boundaries and compute volumes on an unannotated dataset from 60 patients. The dataset includes slice-level start/end annotations for aneurysm but no pixel-wise aorta segmentations. <b>Method:</b> Our method utilizes a pre-trained UNet to automatically locate the aorta, employs SAM2 to track the aorta through vascular irregularities such as aneurysms down to the iliac bifurcation, and finally uses a Long Short-Term Memory (LSTM) network or expert system to identify the beginning and end points of the aneurysm within the aorta. <b>Results:</b> Despite no manual aorta segmentation, our approach achieves promising accuracy, predicting the aneurysm start point with an <i>R</i><sup>2</sup> score of 71%, the end point with an <i>R</i><sup>2</sup> score of 76%, and the volume with an <i>R</i><sup>2</sup> score of 92%. <b>Conclusions:</b> This technique has the potential to facilitate large-scale aneurysm analysis and improve clinical decision-making by reducing dependence on annotated datasets.
format Article
id doaj-art-0681bbb3a89e42628a3e33fdaa773a4f
institution Kabale University
issn 2075-4418
language English
publishDate 2025-07-01
publisher MDPI AG
record_format Article
series Diagnostics
spelling doaj-art-0681bbb3a89e42628a3e33fdaa773a4f2025-08-20T03:32:26ZengMDPI AGDiagnostics2075-44182025-07-011514180410.3390/diagnostics15141804Automated Aneurysm Boundary Detection and Volume Estimation Using Deep LearningAlireza Bagheri Rajeoni0Breanna Pederson1Susan M. Lessner2Homayoun Valafar3Department of Computer Science and Engineering, University of South Carolina, Columbia, SC 29201, USADepartment Health, Sport, and Human Physiology, University of Iowa, Iowa City, IA 52242, USADepartment of Cell Biology and Anatomy, University of South Carolina School of Medicine, Columbia, SC 29209, USADepartment of Computer Science and Engineering, University of South Carolina, Columbia, SC 29201, USA<b>Background/Objective:</b> Precise aneurysm volume measurement offers a transformative edge for risk assessment and treatment planning in clinical settings. Currently, clinical assessments rely heavily on manual review of medical imaging, a process that is time-consuming and prone to inter-observer variability. The widely accepted standard of care primarily focuses on measuring aneurysm diameter at its widest point, providing a limited perspective on aneurysm morphology and lacking efficient methods to measure aneurysm volumes. Yet, volume measurement can offer deeper insight into aneurysm progression and severity. In this study, we propose an automated approach that leverages the strengths of pre-trained neural networks and expert systems to delineate aneurysm boundaries and compute volumes on an unannotated dataset from 60 patients. The dataset includes slice-level start/end annotations for aneurysm but no pixel-wise aorta segmentations. <b>Method:</b> Our method utilizes a pre-trained UNet to automatically locate the aorta, employs SAM2 to track the aorta through vascular irregularities such as aneurysms down to the iliac bifurcation, and finally uses a Long Short-Term Memory (LSTM) network or expert system to identify the beginning and end points of the aneurysm within the aorta. <b>Results:</b> Despite no manual aorta segmentation, our approach achieves promising accuracy, predicting the aneurysm start point with an <i>R</i><sup>2</sup> score of 71%, the end point with an <i>R</i><sup>2</sup> score of 76%, and the volume with an <i>R</i><sup>2</sup> score of 92%. <b>Conclusions:</b> This technique has the potential to facilitate large-scale aneurysm analysis and improve clinical decision-making by reducing dependence on annotated datasets.https://www.mdpi.com/2075-4418/15/14/1804vasculature segmentationdeep learningimage segmentationaneurysmaneurysm boundary detectionaneurysm volume measurement
spellingShingle Alireza Bagheri Rajeoni
Breanna Pederson
Susan M. Lessner
Homayoun Valafar
Automated Aneurysm Boundary Detection and Volume Estimation Using Deep Learning
Diagnostics
vasculature segmentation
deep learning
image segmentation
aneurysm
aneurysm boundary detection
aneurysm volume measurement
title Automated Aneurysm Boundary Detection and Volume Estimation Using Deep Learning
title_full Automated Aneurysm Boundary Detection and Volume Estimation Using Deep Learning
title_fullStr Automated Aneurysm Boundary Detection and Volume Estimation Using Deep Learning
title_full_unstemmed Automated Aneurysm Boundary Detection and Volume Estimation Using Deep Learning
title_short Automated Aneurysm Boundary Detection and Volume Estimation Using Deep Learning
title_sort automated aneurysm boundary detection and volume estimation using deep learning
topic vasculature segmentation
deep learning
image segmentation
aneurysm
aneurysm boundary detection
aneurysm volume measurement
url https://www.mdpi.com/2075-4418/15/14/1804
work_keys_str_mv AT alirezabagherirajeoni automatedaneurysmboundarydetectionandvolumeestimationusingdeeplearning
AT breannapederson automatedaneurysmboundarydetectionandvolumeestimationusingdeeplearning
AT susanmlessner automatedaneurysmboundarydetectionandvolumeestimationusingdeeplearning
AT homayounvalafar automatedaneurysmboundarydetectionandvolumeestimationusingdeeplearning