arXiv 2306.034942023-06eess.IVcs.CV

用于多中心CT血管造影扫描中心房及主动脉血管周围脂肪组织分割的结构不同的神经网络块

Structurally Different Neural Network Blocks for the Segmentation of Atrial and Aortic Perivascular Adipose Tissue in Multi-centre CT Angiography Scans

Mohamed bin Zayed University of Artificial Intelligence（Mohamed bin Zayed大学人工智能学院）；University of Oxford（牛津大学）；Oxford University Hospitals NHS Foundation Trust（牛津大学医院 NHS 基础信托）；Caristo Diagnostics LTD（Caristo诊断有限公司）；University of Leicester（利兹大学）；Royal United Hospitals Bath NHS Foundation Trust（皇家联合医院巴斯 NHS 基础信托）；Oxford University（牛津大学）

Ikboljon Sobirov, Cheng Xie, Muhammad Siddique, Parijat Patel, Kenneth Chan, Thomas Halborg, Christos P. Kotanidis, Zarqaish Fatima, Henry West, Sheena Thomas, … 展开作者

Ikboljon Sobirov, Cheng Xie, Muhammad Siddique, Parijat Patel, Kenneth Chan, Thomas Halborg, Christos P. Kotanidis, Zarqaish Fatima, Henry West, Sheena Thomas, Maria Lyasheva, Donna Alexander, David Adlam, Praveen Rao, Das Indrajeet, Aparna Deshpande, Amrita Bajaj, Jonathan C L Rodrigues, Benjamin J Hudson, Vivek Srivastava, George Krasopoulos, Rana Sayeed, Qiang Zhang, Pete Tomlins, Cheerag Shirodaria, Keith M. Channon, Stefan Neubauer, Charalambos Antoniades, Mohammad Yaqub

备注：: 15 pages, 4 figures, 3 tables

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

提出LegoNet框架，通过交替使用CNN和SwinViT块增强特征学习，用于多中心CTA扫描中血管及周围脂肪组织分割，在外部队列中DSC超0.90，为心血管诊断与预后提供自动化解决方案。

摘要：

自卷积神经网络（CNNs）以及后来的视觉Transformer（ViTs）出现以来，深度学习架构主要依赖于具有不同超参数的相同块类型。我们提出了一种新颖的块交替策略，以利用不同架构设计的互补优势，将结构不同的组件像乐高块一样组装起来。我们引入了LegoNet，这是一个交替使用基于CNN和基于SwinViT的块的深度学习框架，以增强医学图像分割的特征学习。我们研究了LegoNet的三种变体，并将此概念应用于一个此前未被探索的临床问题：从计算机断层扫描血管造影（CTA）扫描中分割内乳动脉（IMA）、主动脉和血管周围脂肪组织（PVAT）。这些PVAT区域已被证明在评估心血管风险和主要临床结果方面具有预后价值。我们在大型数据集上评估了LegoNet，取得了优于其他领先架构的性能。此外，我们在外部测试队列上评估了模型的泛化能力，由专家临床医生纠正模型的分割结果，在各种外部、国际和公共队列中实现了DSC > 0.90。为了进一步验证模型的临床可靠性，我们进行了观察者内和观察者间变异性分析，证明了与人类标注的强一致性。所提出的方法对诊断性心血管管理和早期预后具有重要意义，为临床实践中的血管和血管周围分割及风险评估提供了稳健的自动化解决方案，为个性化医学铺平了道路。

英文摘要：

Since the emergence of convolutional neural networks (CNNs) and, later, vision transformers (ViTs), deep learning architectures have predominantly relied on identical block types with varying hyperparameters. We propose a novel block alternation strategy to leverage the complementary strengths of different architectural designs, assembling structurally distinct components similar to Lego blocks. We introduce LegoNet, a deep learning framework that alternates CNN-based and SwinViT-based blocks to enhance feature learning for medical image segmentation. We investigate three variations of LegoNet and apply this concept to a previously unexplored clinical problem: the segmentation of the internal mammary artery (IMA), aorta, and perivascular adipose tissue (PVAT) from computed tomography angiography (CTA) scans. These PVAT regions have been shown to possess prognostic value in assessing cardiovascular risk and primary clinical outcomes. We evaluate LegoNet on large datasets, achieving superior performance to other leading architectures. Furthermore, we assess the model's generalizability on external testing cohorts, where an expert clinician corrects the model's segmentations, achieving DSC > 0.90 across various external, international, and public cohorts. To further validate the model's clinical reliability, we perform intra- and inter-observer variability analysis, demonstrating strong agreement with human annotations. The proposed methodology has significant implications for diagnostic cardiovascular management and early prognosis, offering a robust, automated solution for vascular and perivascular segmentation and risk assessment in clinical practice, paving the way for personalised medicine.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。