Multi-view hybrid graph convolutional network for volume-to-mesh reconstruction in cardiovascular MRI

IF 11.8 1区 医学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE
Nicolás Gaggion , Benjamin A. Matheson , Yan Xia , Rodrigo Bonazzola , Nishant Ravikumar , Zeike A. Taylor , Diego H. Milone , Alejandro F. Frangi , Enzo Ferrante
{"title":"Multi-view hybrid graph convolutional network for volume-to-mesh reconstruction in cardiovascular MRI","authors":"Nicolás Gaggion ,&nbsp;Benjamin A. Matheson ,&nbsp;Yan Xia ,&nbsp;Rodrigo Bonazzola ,&nbsp;Nishant Ravikumar ,&nbsp;Zeike A. Taylor ,&nbsp;Diego H. Milone ,&nbsp;Alejandro F. Frangi ,&nbsp;Enzo Ferrante","doi":"10.1016/j.media.2025.103630","DOIUrl":null,"url":null,"abstract":"<div><div>Cardiovascular magnetic resonance imaging is emerging as a crucial tool to examine cardiac morphology and function. Essential to this endeavour are anatomical 3D surface and volumetric meshes derived from CMR images, which facilitate computational anatomy studies, biomarker discovery, and in-silico simulations. Traditional approaches typically follow complex multi-step pipelines, first segmenting images and then reconstructing meshes, making them time-consuming and prone to error propagation. In response, we introduce HybridVNet, a novel architecture for direct image-to-mesh extraction seamlessly integrating standard convolutional neural networks with graph convolutions, which we prove can efficiently handle surface and volumetric meshes by encoding them as graph structures. To further enhance accuracy, we propose a multi-view HybridVNet architecture which processes both long axis and short axis CMR, showing that it can increase the performance of cardiac MR mesh generation. Our model combines traditional convolutional networks with variational graph generative models, deep supervision and mesh-specific regularisation. Experiments on a comprehensive dataset from the UK Biobank confirm the potential of HybridVNet to significantly advance cardiac imaging and computational cardiology by efficiently generating high-fidelity meshes from CMR images. Multi-view HybridVNet outperforms the state-of-the-art, achieving improvements of up to <span><math><mo>∼</mo></math></span>27% reduction in Mean Contour Distance (from 1.86 mm to 1.35 mm for the LV Myocardium), up to <span><math><mo>∼</mo></math></span>18% improvement in Hausdorff distance (from 4.74 mm to 3.89 mm, for the LV Endocardium), and up to <span><math><mo>∼</mo></math></span>8% in Dice Coefficient (from 0.78 to 0.84, for the LV Myocardium), highlighting its superior accuracy.</div></div>","PeriodicalId":18328,"journal":{"name":"Medical image analysis","volume":"104 ","pages":"Article 103630"},"PeriodicalIF":11.8000,"publicationDate":"2025-05-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Medical image analysis","FirstCategoryId":"5","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S136184152500177X","RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0

Abstract

Cardiovascular magnetic resonance imaging is emerging as a crucial tool to examine cardiac morphology and function. Essential to this endeavour are anatomical 3D surface and volumetric meshes derived from CMR images, which facilitate computational anatomy studies, biomarker discovery, and in-silico simulations. Traditional approaches typically follow complex multi-step pipelines, first segmenting images and then reconstructing meshes, making them time-consuming and prone to error propagation. In response, we introduce HybridVNet, a novel architecture for direct image-to-mesh extraction seamlessly integrating standard convolutional neural networks with graph convolutions, which we prove can efficiently handle surface and volumetric meshes by encoding them as graph structures. To further enhance accuracy, we propose a multi-view HybridVNet architecture which processes both long axis and short axis CMR, showing that it can increase the performance of cardiac MR mesh generation. Our model combines traditional convolutional networks with variational graph generative models, deep supervision and mesh-specific regularisation. Experiments on a comprehensive dataset from the UK Biobank confirm the potential of HybridVNet to significantly advance cardiac imaging and computational cardiology by efficiently generating high-fidelity meshes from CMR images. Multi-view HybridVNet outperforms the state-of-the-art, achieving improvements of up to 27% reduction in Mean Contour Distance (from 1.86 mm to 1.35 mm for the LV Myocardium), up to 18% improvement in Hausdorff distance (from 4.74 mm to 3.89 mm, for the LV Endocardium), and up to 8% in Dice Coefficient (from 0.78 to 0.84, for the LV Myocardium), highlighting its superior accuracy.
多视图混合图卷积网络用于心血管MRI的体-网格重建
心血管磁共振成像正在成为检查心脏形态和功能的重要工具。这一努力的关键是来自CMR图像的解剖三维表面和体积网格,这有助于计算解剖学研究,生物标志物发现和计算机模拟。传统的方法通常遵循复杂的多步骤管道,首先分割图像,然后重建网格,这使得它们既耗时又容易出错。作为回应,我们引入了HybridVNet,这是一种用于直接图像到网格提取的新架构,将标准卷积神经网络与图卷积无缝集成,我们证明了它可以通过将表面和体积网格编码为图结构来有效地处理它们。为了进一步提高准确性,我们提出了一种同时处理长轴和短轴CMR的多视图HybridVNet架构,表明它可以提高心脏MR网格生成的性能。我们的模型结合了传统的卷积网络与变分图生成模型,深度监督和特定网格的正则化。在英国生物银行的综合数据集上进行的实验证实了HybridVNet的潜力,通过有效地从CMR图像中生成高保真网格,可以显著推进心脏成像和计算心脏病学。多视图HybridVNet优于最先进的技术,平均轮廓距离减少了27%(从1.86 mm到1.35 mm), Hausdorff距离提高了18%(从4.74 mm到3.89 mm,对于左室心内膜),Dice系数提高了8%(从0.78到0.84,对于左室心肌),突出了其优越的准确性。
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
Medical image analysis
Medical image analysis 工程技术-工程:生物医学
CiteScore
22.10
自引率
6.40%
发文量
309
审稿时长
6.6 months
期刊介绍: Medical Image Analysis serves as a platform for sharing new research findings in the realm of medical and biological image analysis, with a focus on applications of computer vision, virtual reality, and robotics to biomedical imaging challenges. The journal prioritizes the publication of high-quality, original papers contributing to the fundamental science of processing, analyzing, and utilizing medical and biological images. It welcomes approaches utilizing biomedical image datasets across all spatial scales, from molecular/cellular imaging to tissue/organ imaging.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:604180095
Book学术官方微信