World of Forms: Deformable Geometric Templates for One-Shot Surface Meshing in Coronary CT Angiography

Rudolf L. M. van Herten, Ioannis Lagogiannis, Jelmer M. Wolterink, Steffen Bruns, Eva R. Meulendijks, Damini Dey, Joris R. de Groot, José P. Henriques, R. Nils Planken, Simone Saitta, Ivana Išgum
{"title":"World of Forms: Deformable Geometric Templates for One-Shot Surface Meshing in Coronary CT Angiography","authors":"Rudolf L. M. van Herten, Ioannis Lagogiannis, Jelmer M. Wolterink, Steffen Bruns, Eva R. Meulendijks, Damini Dey, Joris R. de Groot, José P. Henriques, R. Nils Planken, Simone Saitta, Ivana Išgum","doi":"arxiv-2409.11837","DOIUrl":null,"url":null,"abstract":"Deep learning-based medical image segmentation and surface mesh generation\ntypically involve a sequential pipeline from image to segmentation to meshes,\noften requiring large training datasets while making limited use of prior\ngeometric knowledge. This may lead to topological inconsistencies and\nsuboptimal performance in low-data regimes. To address these challenges, we\npropose a data-efficient deep learning method for direct 3D anatomical object\nsurface meshing using geometric priors. Our approach employs a multi-resolution\ngraph neural network that operates on a prior geometric template which is\ndeformed to fit object boundaries of interest. We show how different templates\nmay be used for the different surface meshing targets, and introduce a novel\nmasked autoencoder pretraining strategy for 3D spherical data. The proposed\nmethod outperforms nnUNet in a one-shot setting for segmentation of the\npericardium, left ventricle (LV) cavity and the LV myocardium. Similarly, the\nmethod outperforms other lumen segmentation operating on multi-planar\nreformatted images. Results further indicate that mesh quality is on par with\nor improves upon marching cubes post-processing of voxel mask predictions,\nwhile remaining flexible in the choice of mesh triangulation prior, thus paving\nthe way for more accurate and topologically consistent 3D medical object\nsurface meshing.","PeriodicalId":501289,"journal":{"name":"arXiv - EE - Image and Video Processing","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - EE - Image and Video Processing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.11837","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

Deep learning-based medical image segmentation and surface mesh generation typically involve a sequential pipeline from image to segmentation to meshes, often requiring large training datasets while making limited use of prior geometric knowledge. This may lead to topological inconsistencies and suboptimal performance in low-data regimes. To address these challenges, we propose a data-efficient deep learning method for direct 3D anatomical object surface meshing using geometric priors. Our approach employs a multi-resolution graph neural network that operates on a prior geometric template which is deformed to fit object boundaries of interest. We show how different templates may be used for the different surface meshing targets, and introduce a novel masked autoencoder pretraining strategy for 3D spherical data. The proposed method outperforms nnUNet in a one-shot setting for segmentation of the pericardium, left ventricle (LV) cavity and the LV myocardium. Similarly, the method outperforms other lumen segmentation operating on multi-planar reformatted images. Results further indicate that mesh quality is on par with or improves upon marching cubes post-processing of voxel mask predictions, while remaining flexible in the choice of mesh triangulation prior, thus paving the way for more accurate and topologically consistent 3D medical object surface meshing.
形态世界:用于冠状动脉 CT 血管造影中一次性表面网格化的可变形几何模板
基于深度学习的医学图像分割和表面网格生成通常涉及从图像到分割再到网格的顺序流水线,往往需要大量的训练数据集,同时对先验几何知识的利用有限。这可能会导致拓扑不一致,以及在低数据量情况下性能不佳。为了应对这些挑战,我们提出了一种数据高效的深度学习方法,利用几何先验知识直接对三维解剖物体表面进行网格划分。我们的方法采用了多分辨率图神经网络,该网络在先验几何模板上运行,模板经过变形以适合感兴趣的物体边界。我们展示了不同的模板如何用于不同的表面网格划分目标,并针对三维球形数据引入了一种新颖的掩码自动编码器预训练策略。在对心包、左心室(LV)腔和左心室心肌进行一次性分割时,所提出的方法优于 nnUNet。同样,该方法在多平面格式图像上的表现也优于其他管腔分割方法。结果进一步表明,网格质量与行进立方体后处理体素掩模预测不相上下,甚至更胜一筹,同时还能灵活选择网格三角先验,从而为更精确、拓扑更一致的三维医学物体表面网格划分铺平了道路。
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
自引率
0.00%
发文量
0
×
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:481959085
Book学术官方微信