Diagnosis and detection of infected tissue of COVID-19 patients based on lung x-ray image using convolutional neural network approaches
Diagnosis and detection of infected tissue of COVID-19 patients based on lung x-ray image using convolutional neural network approaches
复制标题
DOI：
10.1016/j.chaos.2020.110170
复制
发表时间：
2020-11-01
影响因子：
7.8
通讯作者：
Sharifi, Abbas
中科院分区：
文献类型：
--
作者：
Hassantabar, Shayan;Ahmadi, Mohsen;Sharifi, Abbas
COVID-19大流行挑战了世界科学。国际社会试图尽快发现、应用或设计新的方法来诊断和治疗COVID-19患者。目前，用于诊断感染患者的可靠方法是逆转录-聚合酶链反应。该方法昂贵且耗时。因此，设计新颖的方法是重要的。在本文中，我们使用了三种基于深度学习的方法，通过肺部X射线图像来检测和诊断COVID-19患者。对于疾病的诊断，我们提出了两种算法，包括基于图像分形特征的深度神经网络（DNN）和直接使用肺部图像的卷积神经网络（CNN）方法。分类结果表明，所提出的CNN架构具有更高的准确度（93.2%）和灵敏度（96.1%），优于DNN方法，准确度为83.4%，灵敏度为86%。在分割过程中，我们提出了一种CNN架构来找到肺部图像中的感染组织。实验结果表明，该方法能以83.84%的准确率检测出几乎所有的感染区域。这一发现也可用于监测和控制病人从感染区域的增长。(C)2020由Elsevier Ltd.出版
COVID-19 pandemic has challenged the world science. The international community tries to find, apply, or design novel methods for diagnosis and treatment of COVID-19 patients as soon as possible. Currently, a reliable method for the diagnosis of infected patients is a reverse transcription-polymerase chain reaction. The method is expensive and time-consuming. Therefore, designing novel methods is important. In this paper, we used three deep learning-based methods for the detection and diagnosis of COVID-19 patients with the use of X-Ray images of lungs. For the diagnosis of the disease, we presented two algorithms include deep neural network (DNN) on the fractal feature of images and convolutional neural network (CNN) methods with the use of the lung images, directly. Results classification shows that the presented CNN architecture with higher accuracy (93.2%) and sensitivity (96.1%) is outperforming than the DNN method with an accuracy of 83.4% and sensitivity of 86%. In the segmentation process, we presented a CNN architecture to find infected tissue in lung images. Results show that the presented method can almost detect infected regions with high accuracy of 83.84%. This finding also can be used to monitor and control patients from infected region growth. (C) 2020 Published by Elsevier Ltd.