Multiscale dense convolutional neural network for DSA cerebrovascular segmentation
Multiscale dense convolutional neural network for DSA cerebrovascular segmentation
复制标题
用于 DSA 脑血管分割的多尺度密集卷积神经网络
DOI：
10.1016/j.neucom.2019.10.035
复制
发表时间：
2020-01-15
期刊：
影响因子：
6
通讯作者：
Liu, Lei
中科院分区：
文献类型：
--
作者：
Meng, Cai;Sun, Kai;Liu, Lei
在数字减影血管造影（DSA）图像中，准确的脑血管分割是一个不可缺少的步骤，可以帮助医生正确估计脑血管病变的程度，避免误诊。由于脑血管结构的复杂性和DSA中造影剂分布的不均匀性，自动分割是临床诊断中具有挑战性的任务。近年来，深度卷积神经网络（CNN）已经超越了最先进的方法，并在医学图像分割方面显示出巨大的潜力。本文提出了一种基于CNN的分割框架MDCNN，用于自动分割DSA图像中的脑血管。受U-net的启发，该MDCNN结构被设计为编码器-解码器架构。考虑到脑血管中血管直径的多样性，我们定义了一个多尺度模型来分割不同直径的脑血管。同时，我们重新设计了编码器和解码器之间的跳过连接，以利用编码器级的更多功能。为了提高提取高级特征的能力，引入了改进的密集块。在简化参数方面，我们参考了深度监督的想法，使修剪成为可能。该框架在DCVessel（我们实验室制作的DSA脑血管数据集）上进行了测试。我们提出的方法在F1评分、准确性（Acc）、灵敏度（Sen）、特异性（Spe）和AUC方面分别达到0.8813、0.9784、0.8775、0.9886、0.9944，优于现有的方法。同时，我们在基准视网膜血管数据集DRIVE上评估了我们的MDCNN框架。实验结果表明，所提出的MDCNN模型是可扩展的各种血管分割任务。(C)2019 Elsevier B. V.版权所有。
Accurate cerebrovascular segmentation in Digital Subtraction Angiography (DSA) image, as an indispensable step, can help doctors appropriately estimate the degree of cerebrovascular lesions to avoid misdiagnosis. Because of the complexity of cerebrovascular structure and the uneven distribution of contrast media in DSA, automatic segmentation is a challenging task in clinical diagnosis. In recent years, deep convolutional neural networks (CNN) have outperformed the state-of-art methods and shown great potential for medical image segmentation. This paper proposes a CNN-based segmentation framework Multiscale Dense CNN (MDCNN) to automatically segment cerebral vessel in DSA images. Inspired by U-net, this proposed MDCNN structure is designed as encoder-decoder architecture. Considering that the diameters of blood vessel in cerebrovascular are various, we define a multiscale module to segment cerebral vessel with different diameters. Meanwhile, we redesign the skip connections between encoder and decoder stage to utilize more features from encoder stage. To improve the capability of extracting high-level features, improved dense blocks are introduced. In terms of simplifying parameters, we refer to the idea of deep supervision to make pruning possible. The proposed framework is tested on DCVessel (a DSA cerebrovascular dataset made by our lab). Our proposed method reaches 0.8813, 0.9784, 0.8775, 0.9886, 0.9944 in F1 score, accuracy (Acc), sensitivity (Sen), specificity (Spe) and AUC respectively, outperforming the state-of-art methods. Meanwhile, we evaluate our MDCNN framework on benchmark retinal vessel dataset DRIVE. The promising experiment results demonstrate that proposed MDCNN model is extensible for various vessel segmentation tasks. (C) 2019 Elsevier B.V. All rights reserved.