Retinal vessel segmentation of color fundus images using multiscale convolutional neural network with an improved cross-entropy loss function
Retinal vessel segmentation of color fundus images using multiscale convolutional neural network with an improved cross-entropy loss function
复制标题
使用具有改进的交叉熵损失函数的多尺度卷积神经网络对彩色眼底图像进行视网膜血管分割
DOI：
10.1016/j.neucom.2018.05.011
复制
发表时间：
2018-10-02
期刊：
影响因子：
6
通讯作者：
Gao, Xieping
中科院分区：
文献类型：
--
作者：
Hu, Kai;Zhang, Zhenzhen;Gao, Xieping
眼底图像的视网膜血管分析是筛查和诊断相关疾病不可缺少的方法。提出了一种基于卷积神经网络和全连接条件随机场的眼底血管分割方法。分割过程主要分为两个步骤。首先，提出了一种具有改进的交叉熵损失函数的多尺度CNN架构，以产生从图像到图像的概率图。我们通过结合各个中间层的特征图来构建多尺度网络，以学习视网膜血管的更多细节信息。同时，我们提出的交叉熵损失函数忽略了相对容易的样本的最小损失，以便更多地关注学习困难的样本。其次，利用CRFs对眼底图像进行二值化分割，充分考虑了眼底图像中各像素点之间的相互作用，利用了更多的空间背景信息，得到最终的二值化分割结果。所提出的方法的有效性已经在两个公共数据集上进行了评估，即，DRIVE和STARE与11种最先进的方法进行比较，包括5种基于深度学习的方法。结果表明，我们的方法可以检测到更多的微小血管和更精确的边缘定位。(C)2018爱思唯尔B.V.保留所有权利。
Retinal vessel analysis of fundus images is an indispensable method for the screening and diagnosis of related diseases. In this paper, we propose a novel retinal vessel segmentation method of the fundus images based on convolutional neural network (CNN) and fully connected conditional random fields (CRFs). The segmentation process is mainly divided into two steps. Firstly, a multiscale CNN architecture with an improved cross-entropy loss function is proposed to produce the probability map from image to image. We construct the multiscale network by combining the feature map of each middle layer to learn more detail information of the retinal vessels. Meanwhile, our proposed cross-entropy loss function ignores the slightest loss of relatively easy samples in order to take more attention to learn the hard examples. Secondly, CRFs is applied to get the final binary segmentation result which makes use of more spatial context information by taking into account the interactions among all of the pixels in the fundus images. The effectiveness of the proposed method has been evaluated on two public datasets, i.g., DRIVE and STARE with comparisons against eleven state-of-the-art approaches including five deep learning based methods. Results show that our method allows for detection of more tiny blood vessels and more precise locating of the edges. (C) 2018 Elsevier B.V. All rights reserved.