Development and testing of a deep learning-based strategy for scar segmentation on CMR-LGE images
Development and testing of a deep learning-based strategy for scar segmentation on CMR-LGE images
复制标题
DOI：
10.1007/s10334-018-0718-4
复制
发表时间：
2019-04-01
影响因子：
2.3
通讯作者：
Caiani, Enrico Gianluca
中科院分区：
文献类型：
--
作者：
Moccia, Sara;Banali, Riccardo;Caiani, Enrico Gianluca
本文的目的是研究使用全卷积神经网络（FCNN）分割疤痕组织在左心室的心脏磁共振与晚期钆增强（CMR-LGE）images.MethodsA成功的FCNN在文献中（ENet）进行了修改和训练，以提供疤痕组织分割。研究了两种分割协议（协议1和协议2），后者将瘢痕分割搜索区域限制在左心室心肌组织区域。回顾性分析了30例缺血性心脏病患者的CMR-LGE，共250张图像，在瘢痕尺寸和位置方面呈现出高度变异性。分割结果进行了评估，对手动疤痕组织跟踪使用一个病人的交叉validation.ResultsProtocol 2优于协议1显着（p值< 0.05），中位灵敏度和Dice相似系数分别为88.07% [四分位距（IQR）18.84%]和71.25%（IQR 31.82%），分别。DiscussionBoth分割协议能够检测到疤痕组织的CMR-LGE图像，但更高的性能时，实现了限制搜索区域心肌区域。本文的研究结果代表了使用FCNN从CMR-LGE图像中分割无活力瘢痕组织的一个令人鼓舞的起点。
ObjectiveThe aim of this paper is to investigate the use of fully convolutional neural networks (FCNNs) to segment scar tissue in the left ventricle from cardiac magnetic resonance with late gadolinium enhancement (CMR-LGE) images.MethodsA successful FCNN in the literature (the ENet) was modified and trained to provide scar-tissue segmentation. Two segmentation protocols (Protocol 1 and Protocol 2) were investigated, the latter limiting the scar-segmentation search area to the left ventricular myocardial tissue region. CMR-LGE from 30 patients with ischemic-heart disease were retrospectively analyzed, for a total of 250 images, presenting high variability in terms of scar dimension and location. Segmentation results were assessed against manual scar-tissue tracing using one-patient-out cross validation.ResultsProtocol 2 outperformed Protocol 1 significantly (p value < 0.05), with median sensitivity and Dice similarity coefficient equal to 88.07% [inter-quartile range (IQR) 18.84%] and 71.25% (IQR 31.82%), respectively.DiscussionBoth segmentation protocols were able to detect scar tissues in the CMR-LGE images but higher performance was achieved when limiting the search area to the myocardial region. The findings of this paper represent an encouraging starting point for the use of FCNNs for the segmentation of nonviable scar tissue from CMR-LGE images.