Hybrid-COVID: a novel hybrid 2D/3D CNN based on cross-domain adaptation approach for COVID-19 screening from chest X-ray images.
Hybrid-COVID: a novel hybrid 2D/3D CNN based on cross-domain adaptation approach for COVID-19 screening from chest X-ray images.
复制标题
DOI：
10.1007/s13246-020-00957-1
复制
发表时间：
2020-12
影响因子：
4.4
通讯作者：
Mtibaa A
中科院分区：
文献类型：
--
作者：
Bayoudh K;Hamdaoui F;Mtibaa A
新型冠状病毒病（COVID-19）于2019年12月底首次出现，目前仍在世界大多数国家迅速传播。呼吸道感染主要发生在大多数接受COVID-19治疗的患者中。鉴于COVID-19病例数量不断增加，对诊断工具的需求在早期阶段识别COVID-19感染至关重要。几十年来，胸部X射线（CXR）技术已经证明了其准确检测呼吸系统疾病的能力。最近，随着COVID-19 CXR扫描的可用性，深度学习算法在医疗保健竞技场中发挥了关键作用，允许放射科医生从他们的CXR图像中识别COVID-19患者。然而，最近研究中报告的大多数COVID-19筛查方法都是基于2D卷积神经网络（CNN）。虽然3D CNN与2D CNN相比能够捕获上下文信息，但由于其增加的计算成本（即需要更多的额外内存和更多的计算能力），它们的使用受到限制。在这项研究中，开发了一种基于迁移学习的混合2D/3D CNN架构，用于使用CXR进行COVID-19筛查。所提出的架构包括一个预先训练的深度模型（VGG 16）和一个浅层3D CNN，结合一个深度可分离的卷积层和一个空间金字塔池化模块（SPP）。具体来说，深度可分离卷积有助于保留有用的特征，同时减少模型的计算负担。SPP模块被设计为从中间表示中提取多级表示。实验结果表明，该框架在收集的数据集（3个待预测类别：COVID-19，肺炎和正常）上进行评估时可以实现合理的性能。其敏感性为98.33%，特异性为98.68%，总准确性为96.91%
The novel Coronavirus disease (COVID-19), which first appeared at the end of December 2019, continues to spread rapidly in most countries of the world. Respiratory infections occur primarily in the majority of patients treated with COVID-19. In light of the growing number of COVID-19 cases, the need for diagnostic tools to identify COVID-19 infection at early stages is of vital importance. For decades, chest X-ray (CXR) technologies have proven their ability to accurately detect respiratory diseases. More recently, with the availability of COVID-19 CXR scans, deep learning algorithms have played a critical role in the healthcare arena by allowing radiologists to recognize COVID-19 patients from their CXR images. However, the majority of screening methods for COVID-19 reported in recent studies are based on 2D convolutional neural networks (CNNs). Although 3D CNNs are capable of capturing contextual information compared to their 2D counterparts, their use is limited due to their increased computational cost (i.e. requires much extra memory and much more computing power). In this study, a transfer learning-based hybrid 2D/3D CNN architecture for COVID-19 screening using CXRs has been developed. The proposed architecture consists of the incorporation of a pre-trained deep model (VGG16) and a shallow 3D CNN, combined with a depth-wise separable convolution layer and a spatial pyramid pooling module (SPP). Specifically, the depth-wise separable convolution helps to preserve the useful features while reducing the computational burden of the model. The SPP module is designed to extract multi-level representations from intermediate ones. Experimental results show that the proposed framework can achieve reasonable performances when evaluated on a collected dataset (3 classes to be predicted: COVID-19, Pneumonia, and Normal). Notably, it achieved a sensitivity of 98.33%, a specificity of 98.68% and an overall accuracy of 96.91%
登录
查看更多内容
DOI：
10.1109/tpami.2012.59
发表时间：
2013-01-01
影响因子：
23.6
作者：
Ji, Shuiwang;Xu, Wei;Yu, Kai
通讯作者：
Yu, Kai
影响因子：
6.1
作者：
Khan, Asif Iqbal;Shah, Junaid Latief;Bhat, Mohammad Mudasir
通讯作者：
Bhat, Mohammad Mudasir
DOI：
10.1007/s10489-020-01714-3
发表时间：
2020-04-22
期刊：
Applied Intelligence (Dordrecht, Netherlands)
影响因子：
--
作者：
Butt C;Gill J;Chun D;Babu BA
通讯作者：
Babu BA
影响因子：
12
作者：
Khan, Asifullah;Sohail, Anabia;Qureshi, Aqsa Saeed
通讯作者：
Qureshi, Aqsa Saeed
影响因子：
4.4
作者：
Apostolopoulos, Ioannis D.;Mpesiana, Tzani A.
通讯作者：
Mpesiana, Tzani A.