期刊
JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY
卷 37, 期 2, 页码 330-343出版社
SCIENCE PRESS
DOI: 10.1007/s11390-020-0679-8
关键词
pneumonia; COVID-19; convolutional neural network; AlexNet; deep learning
资金
- Royal Society International Exchanges Cost Share Award of UK [RP202G0230]
- Medical Research Council Confidence in Concept Award of UK [MC PC 17171]
- Hope Foundation for Cancer Research of UK [RM60G0680]
- British Heart Foundation Accelerator Award of UK [AA/18/3/34220]
- Sino-UK Industrial Fund [RP202G0289]
- Global Challenges Research Fund (GCRF) of UK [P202PF11]
- Fundamental Research Funds for the Central Universities of China [CDLS-2020-03]
- Key Laboratory of Child Development and Learning Science (Southeast University), Ministry of Education of China, Henan Key Research and Development Project of China [182102310629]
- National Natural Science Foundation of China [U19B2032, 61772511]
In this study, a deep learning network-based framework for COVID-19 diagnosis is proposed. By improving AlexNet and introducing three classifiers, three novel models are obtained. Among them, DC-Net-R performs the best on a private dataset and outperforms other existing algorithms.
COVID-19 is a contagious infection that has severe effects on the global economy and our daily life. Accurate diagnosis of COVID-19 is of importance for consultants, patients, and radiologists. In this study, we use the deep learning network AlexNet as the backbone, and enhance it with the following two aspects: 1) adding batch normalization to help accelerate the training, reducing the internal covariance shift; 2) replacing the fully connected layer in AlexNet with three classifiers: SNN, ELM, and RVFL. Therefore, we have three novel models from the deep COVID network (DC-Net) framework, which are named DC-Net-S, DC-Net-E, and DC-Net-R, respectively. After comparison, we find the proposed DC-Net-R achieves an average accuracy of 90.91% on a private dataset (available upon email request) comprising of 296 images while the specificity reaches 96.13%, and has the best performance among all three proposed classifiers. In addition, we show that our DC-Net-R also performs much better than other existing algorithms in the literature.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据