Fully Automated Aortic Root Localization and Tilt Alignment in Cardiac Computed Tomography

Journal of the Society for Cardiovascular Angiography & Interventions Pub Date : 2025-08-01 DOI:10.1016/j.jscai.2025.103716

Elham Mahmoudi MD, MPH , Vinayak Nagaraja MD , Mohamad Sarraf MD , Paul Friedman MD , Mohamad Alkhouli MD , Mackram F. Eleid MD , Mandeep Singh MD , Zachi I. Attia PhD , Joseph D. Sobek MSc , Mohammadreza Naderian MD, MPH , Fred Nugen PhD , Bardia Khosravi MD, MPH, MHPE , Sanaz Vahdati MD , Bradley J. Erickson MD, PhD

{"title":"Fully Automated Aortic Root Localization and Tilt Alignment in Cardiac Computed Tomography","authors":"Elham Mahmoudi MD, MPH , Vinayak Nagaraja MD , Mohamad Sarraf MD , Paul Friedman MD , Mohamad Alkhouli MD , Mackram F. Eleid MD , Mandeep Singh MD , Zachi I. Attia PhD , Joseph D. Sobek MSc , Mohammadreza Naderian MD, MPH , Fred Nugen PhD , Bardia Khosravi MD, MPH, MHPE , Sanaz Vahdati MD , Bradley J. Erickson MD, PhD","doi":"10.1016/j.jscai.2025.103716","DOIUrl":null,"url":null,"abstract":"<div><h3>Background</h3><div>Automated analysis of cardiac computed tomography (CCT) studies may help in personalized management and outcome prediction of patients undergoing transcatheter aortic valve replacement (TAVR). The current methods are often preceded by a manual selection of the region of interest. To address this limitation, this study aims to develop an object-oriented aortic root detection pipeline.</div></div><div><h3>Methods</h3><div>All consecutive patients who underwent CCT for TAVR procedure, from January to July 2023 at our center, were retrospectively collected. Patients with previous prosthesis or permanent pacemaker were excluded. Baseline bounding box annotations were performed by a single expert, and tilt angle measurements were performed by 2 for interobserver comparison. A pretrained convolutional neural network was used for aortic root detection, and its performance was evaluated by recall, precision, F1, average precision at an intersection over union overlap of 50% and mean average precision (mAP) 50%-95% on 100 unseen test set. For tilt alignment, intensity thresholding, connected component, and principal component analyses were proposed. Results were evaluated by Bland-Altman comparison.</div></div><div><h3>Results</h3><div>Of the 228 TAVR patients with preprocedural CCT, 179 were eligible, and their axial contrast-enhanced CCTs could be retrieved successfully; 100 CCTs were assigned to the test set, and the remaining to the training and validation using a 4:1 split. The model detected the aortic root with recall, precision, and F1 score of 99.0%, for all 3; mAP50 of 99.5%; and mAP50-95 of 60.4%. The tilt prediction algorithm had a mean error of 7.9° (Q1-Q3, −5.3° to 21.1°) compared with 3.3° (Q1-Q3, −6.7° to 13.4°) interobserver difference.</div></div><div><h3>Conclusions</h3><div>This study demonstrates the robust performance of a fully automated pipeline for aortic root detection and analysis of key features in pre-TAVR CCTs. Further prospective studies are required for clinical developments.</div></div>","PeriodicalId":73990,"journal":{"name":"Journal of the Society for Cardiovascular Angiography & Interventions","volume":"4 8","pages":"Article 103716"},"PeriodicalIF":0.0000,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of the Society for Cardiovascular Angiography & Interventions","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2772930325011585","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}

引用次数: 0

Abstract

Background

Automated analysis of cardiac computed tomography (CCT) studies may help in personalized management and outcome prediction of patients undergoing transcatheter aortic valve replacement (TAVR). The current methods are often preceded by a manual selection of the region of interest. To address this limitation, this study aims to develop an object-oriented aortic root detection pipeline.

Methods

All consecutive patients who underwent CCT for TAVR procedure, from January to July 2023 at our center, were retrospectively collected. Patients with previous prosthesis or permanent pacemaker were excluded. Baseline bounding box annotations were performed by a single expert, and tilt angle measurements were performed by 2 for interobserver comparison. A pretrained convolutional neural network was used for aortic root detection, and its performance was evaluated by recall, precision, F1, average precision at an intersection over union overlap of 50% and mean average precision (mAP) 50%-95% on 100 unseen test set. For tilt alignment, intensity thresholding, connected component, and principal component analyses were proposed. Results were evaluated by Bland-Altman comparison.

Results

Of the 228 TAVR patients with preprocedural CCT, 179 were eligible, and their axial contrast-enhanced CCTs could be retrieved successfully; 100 CCTs were assigned to the test set, and the remaining to the training and validation using a 4:1 split. The model detected the aortic root with recall, precision, and F1 score of 99.0%, for all 3; mAP50 of 99.5%; and mAP50-95 of 60.4%. The tilt prediction algorithm had a mean error of 7.9° (Q1-Q3, −5.3° to 21.1°) compared with 3.3° (Q1-Q3, −6.7° to 13.4°) interobserver difference.

Conclusions

This study demonstrates the robust performance of a fully automated pipeline for aortic root detection and analysis of key features in pre-TAVR CCTs. Further prospective studies are required for clinical developments.

查看原文本刊更多论文

在心脏计算机断层扫描中全自动主动脉根部定位和倾斜对齐

背景：心脏计算机断层扫描（CCT）研究的自动化分析可能有助于经导管主动脉瓣置换术（TAVR）患者的个性化管理和预后预测。目前的方法通常是先手动选择感兴趣的区域。为了解决这一局限性，本研究旨在开发一种面向对象的主动脉根部检测管道。方法回顾性收集2013年1月至7月在本中心连续行CCT治疗TAVR的患者。排除既往使用假体或永久性起搏器的患者。基线边界框注释由一名专家执行，倾斜角测量由两名专家执行，用于观察者之间的比较。采用预训练的卷积神经网络进行主动脉根部检测，在100个未见的测试集上，通过召回率、精度、F1、相交处的平均精度为50%和平均精度（mAP）为50% ~ 95%来评价其性能。对于倾斜定位，提出了强度阈值、连接成分和主成分分析。采用Bland-Altman比较评价结果。结果228例TAVR患者行术前CCT，其中179例符合条件，均可成功检索轴向增强CCT；100个cct被分配给测试集，其余的被分配给训练和验证，使用4:1的分割。模型检测主动脉根部的查全率、查准率和F1评分均为99.0%；mAP50为99.5%；mAP50-95为60.4%。倾斜预测算法的平均误差为7.9°（Q1-Q3,−5.3°至21.1°），而观察者间误差为3.3°（Q1-Q3,−6.7°至13.4°）。本研究证明了全自动管道在主动脉根部检测和tavr前cct关键特征分析中的强大性能。临床发展需要进一步的前瞻性研究。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊