臺灣博碩士論文加值系統

目的 : 許多研究人員已設計用於檢測新冠肺炎的深度學習系統，但大多數的系統無法檢測出新冠肺炎與其他肺炎的差異，進而增加了現有X光影像檢測新冠肺炎(COVID-19)與其他病毒型肺炎的難度。因此，本研究設計了一種基於深度學習專用於COVID-19的快速自動檢測網路框架Fast-CovNet。
材料與方法 : 透過COVID-19 RADIOGRAPHY DATABASE數據集(3886組影像)，使用具有醫師標記的胸部X光影像識別感染COVID-19的患者。系統將影像分為三類：COVID-19陽性、其他病毒型肺炎和無感染，研究設計了一種基於卷積神經網路(Convolutional Neural Network, CNN)的Fast-CovNet，並與兩種深度遷移學習演算法(VGG-16、ResNet50)進行比較。
結果 : 研究以十次交叉驗證(Holdout cross validation)評估了COVID-19自動檢測的三種演算法，結果如下：1. Fast-CovNet訓練至檢測的速度為207.64±11.60秒，AUC值最佳為0.98。2. VGG16訓練至檢測的速度為359.75±18.74秒，AUC值最佳為0.98。3. ResNet50訓練至檢測的速度為574.98±17.51秒，AUC值最佳為0.94。
結論 : 本研究提出了一種基於CNN的Fast-CovNet框架，可快速協助進行胸部X光影像自動檢測新冠肺炎，有助於輔助診斷以識別COVID-19與其他病毒型的患者，減少胸部X光影像判讀為偽陰性的風險。

Purpose: Many researchers have designed deep learning systems to detect new coronary pneumonia,But most systems cannot detect the difference between new coronary pneumonia and other pneumonias,this further increases the difficulty of existing X-ray imaging to detect new coronary pneumonia (COVID-19) and other viral pneumonias.Therefore,In this research, we designed a fast and automatic detection network framework Fast-CovNet dedicated to COVID-19 based on deep learning.
Materials and Methods : Through the COVID-19 RADIOGRAPHY DATABASE data set (3886 sets of images), use chest X-ray images marked by physicians to identify patients infected with COVID-19. The system divides the images into three categories: COVID-19 positive, other viral pneumonia and no infection. The research designed a Fast-CovNet based on Convolutional Neural Network (CNN) ,and compare with two deep transfer learning algorithms (VGG-16, ResNet50).
Results: The study evaluated three algorithms for automatic detection of COVID-19 with ten times of cross validation (Holdout cross validation) ,The results are as follows: 1. The training speed of Fast-CovNet to detection is 207.64±11.60 seconds, and the best AUC is 0.98. 2. The speed from training to detection of VGG16 is 359.75±18.74 seconds, and the best AUC is 0.98. 3. The speed from training to detection of ResNet50 is 574.98±17.51 seconds, and the best AUC is 0.94.
Conclusion : This research proposes a Fast-CovNet framework based on CNN, which can quickly assist in the automatic detection of new coronary pneumonia in chest X-ray images,helps to assist in diagnosis to identify patients with COVID-19 and other virus types,reduce the risk of false negatives in chest X-ray images.

摘要 i
Abstract ii
致謝 iii
目錄 iv
表目錄 v
圖目錄 vi
符號縮寫 vii
符號說明 viii
第一章 緒論 1
1.1 動機 1
1.2 目的 2
1.3 相關文獻探討 3
第二章 材料與方法 6
2.1 前言 6
2.2 胸部X光影像數據 8
2.3 影像前處理 8
2.4 監督式學習 9
2.5 卷積神經網路 10
2.6 激活函數(Activate function) 12
2.7 VGG-16 13
2.8 ResNet 50 20
2.9 Fast-CovNet網路架構 21
2.10 評估指標 24
第三章 結果 25
3.1 前言 25
3.2 演算法性能評估 25
第四章 討論 33
4.1 前言 33
4.2 比較人工以及AI 檢測Covid-19 34
4.3 Fast-CovNet與現有文獻結果評估 35
第五章 結論 37
參考文獻 38
自傳 46

[1]H. Lu, C. W. Stratton, and Y. W. J. J. o. m. v. Tang, "Outbreak of pneumonia of unknown etiology in Wuhan, China: The mystery and the miracle," vol. 92, no. 4, p. 401, 2020.
[2]A. E. Gorbalenya, S. C. Baker, R. S. Baric, R. J. de Groot, C. Drosten, A. A. Gulyaeva, B. L. Haagmans, C. Lauber, A. M. Leontovich, and B. W. J. B. Neuman, "Severe acute respiratory syndrome-related coronavirus: The species and its viruses–a statement of the Coronavirus Study Group," 2020.
[3]C. Huang, Y. Wang, X. Li, L. Ren, J. Zhao, Y. Hu, L. Zhang, G. Fan, J. Xu, and X. J. T. l. Gu, "Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China," vol. 395, no. 10223, pp. 497-506, 2020.
[4]X. Wang, X. Deng, Q. Fu, Q. Zhou, J. Feng, H. Ma, W. Liu, and C. J. I. t. o. m. i. Zheng, "A weakly-supervised framework for COVID-19 classification and lesion localization from chest CT," vol. 39, no. 8, pp. 2615-2625, 2020.
[5]G. Luengo-Alonso, F. G.-S. Pérez-Tabernero, M. Tovar-Bazaga, J. M. Arguello-Cuenca, and E. J. I. o. Calvo, "Critical adjustments in a department of orthopaedics through the COVID-19 pandemic," vol. 44, pp. 1557-1564, 2020.
[6]T. E. J. B. Knight and Biobanking, "Severe acute respiratory syndrome coronavirus 2 and coronavirus disease 2019: a clinical overview and primer," vol. 18, no. 6, pp. 492-502, 2020.
[7]Y. Wang, M. Hu, Q. Li, X.-P. Zhang, G. Zhai, and N. J. a. p. a. Yao, "Abnormal respiratory patterns classifier may contribute to large-scale screening of people infected with COVID-19 in an accurate and unobtrusive manner," 2020.
[8]J. Chen, L. Wu, J. Zhang, L. Zhang, D. Gong, Y. Zhao, Q. Chen, S. Huang, M. Yang, and X. J. S. r. Yang, "Deep learning-based model for detecting 2019 novel coronavirus pneumonia on high-resolution computed tomography," vol. 10, no. 1, pp. 1-11, 2020.
[9]K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770-778.
[10]N. C. Codella, Q.-B. Nguyen, S. Pankanti, D. A. Gutman, B. Helba, A. C. Halpern, J. R. J. I. J. o. R. Smith, and Development, "Deep learning ensembles for melanoma recognition in dermoscopy images," vol. 61, no. 4/5, pp. 5: 1-5: 15, 2017.
[11]A. Esteva, B. Kuprel, R. A. Novoa, J. Ko, S. M. Swetter, H. M. Blau, and S. J. n. Thrun, "Dermatologist-level classification of skin cancer with deep neural networks," vol. 542, no. 7639, pp. 115-118, 2017.
[12]Y. Dong, Z. Jiang, H. Shen, W. D. Pan, L. A. Williams, V. V. Reddy, W. H. Benjamin, and A. W. Bryan, "Evaluations of deep convolutional neural networks for automatic identification of malaria infected cells," in 2017 IEEE EMBS international conference on biomedical & health informatics (BHI), 2017, pp. 101-104: IEEE.
[13]U. R. Acharya, S. L. Oh, Y. Hagiwara, J. H. Tan, M. Adam, A. Gertych, R. J. C. i. b. San Tan, and medicine, "A deep convolutional neural network model to classify heartbeats," vol. 89, pp. 389-396, 2017.
[14]A. Y. Hannun, P. Rajpurkar, M. Haghpanahi, G. H. Tison, C. Bourn, M. P. Turakhia, and A. Y. J. N. m. Ng, "Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network," vol. 25, no. 1, pp. 65-69, 2019.
[15]Ö. Yıldırım, P. Pławiak, R.-S. Tan, U. R. J. C. i. b. Acharya, and medicine, "Arrhythmia detection using deep convolutional neural network with long duration ECG signals," vol. 102, pp. 411-420, 2018.
[16]G. Gaál, B. Maga, and A. J. a. p. a. Lukács, "Attention u-net based adversarial architectures for chest x-ray lung segmentation," 2020.
[17]J. C. Souza, J. O. B. Diniz, J. L. Ferreira, G. L. F. da Silva, A. C. Silva, A. C. J. C. m. de Paiva, and p. i. biomedicine, "An automatic method for lung segmentation and reconstruction in chest X-ray using deep neural networks," vol. 177, pp. 285-296, 2019.
[18]J. H. Tan, H. Fujita, S. Sivaprasad, S. V. Bhandary, A. K. Rao, K. C. Chua, and U. R. J. I. s. Acharya, "Automated segmentation of exudates, haemorrhages, microaneurysms using single convolutional neural network," vol. 420, pp. 66-76, 2017.
[19]Y. Celik, M. Talo, O. Yildirim, M. Karabatak, and U. R. J. P. R. L. Acharya, "Automated invasive ductal carcinoma detection based using deep transfer learning with whole-slide images," vol. 133, pp. 232-239, 2020.
[20]A. Cruz-Roa, A. Basavanhally, F. González, H. Gilmore, M. Feldman, S. Ganesan, N. Shih, J. Tomaszewski, and A. Madabhushi, "Automatic detection of invasive ductal carcinoma in whole slide images with convolutional neural networks," in Medical Imaging 2014: Digital Pathology, 2014, vol. 9041, p. 904103: International Society for Optics and Photonics.
[21]P. Rajpurkar, J. Irvin, K. Zhu, B. Yang, H. Mehta, T. Duan, D. Ding, A. Bagul, C. Langlotz, and K. J. a. p. a. Shpanskaya, "Chexnet: Radiologist-level pneumonia detection on chest x-rays with deep learning," 2017.
[22]M. Talo, O. Yildirim, U. B. Baloglu, G. Aydin, U. R. J. C. M. I. Acharya, and Graphics, "Convolutional neural networks for multi-class brain disease detection using MRI images," vol. 78, p. 101673, 2019.
[23]X. Xu, X. Jiang, C. Ma, P. Du, X. Li, S. Lv, L. Yu, Q. Ni, Y. Chen, and J. J. E. Su, "A deep learning system to screen novel coronavirus disease 2019 pneumonia," vol. 6, no. 10, pp. 1122-1129, 2020.
[24]Z. Han, B. Wei, Y. Hong, T. Li, J. Cong, X. Zhu, H. Wei, and W. J. I. t. o. m. i. Zhang, "Accurate screening of COVID-19 using attention-based deep 3D multiple instance learning," vol. 39, no. 8, pp. 2584-2594, 2020.
[25]A. Mishal, R. Saravanan, S. S. Atchitha, K. Santhiya, M. Rithika, S. S. Menaka, and T. J. H. Thiruvalluvan, "A Review of Corona Virus Disease-2019," vol. 4, no. 7, 2020.
[26]S. Wang, B. Kang, J. Ma, X. Zeng, M. Xiao, J. Guo, M. Cai, J. Yang, Y. Li, and X. J. E. r. Meng, "A deep learning algorithm using CT images to screen for Corona Virus Disease (COVID-19)," pp. 1-9, 2021.
[27]H. Kang, L. Xia, F. Yan, Z. Wan, F. Shi, H. Yuan, H. Jiang, D. Wu, H. Sui, and C. J. I. t. o. m. i. Zhang, "Diagnosis of coronavirus disease 2019 (COVID-19) with structured latent multi-view representation learning," vol. 39, no. 8, pp. 2606-2614, 2020.
[28]B. Ghoshal and A. J. a. p. a. Tucker, "Estimating uncertainty and interpretability in deep learning for coronavirus (COVID-19) detection," 2020.
[29]X. Bai, C. Fang, Y. Zhou, S. Bai, Z. Liu, L. Xia, Q. Chen, Y. Xu, T. Xia, and S. Gong, "Predicting COVID-19 malignant progression with AI techniques," 2020.
[30]S. Jin, B. Wang, H. Xu, C. Luo, L. Wei, W. Zhao, X. Hou, W. Ma, Z. Xu, and Z. J. M. Zheng, "AI-assisted CT imaging analysis for COVID-19 screening: Building and deploying a medical AI system in four weeks," 2020.
[31]C. Jin, W. Chen, Y. Cao, Z. Xu, Z. Tan, X. Zhang, L. Deng, C. Zheng, J. Zhou, and H. J. N. c. Shi, "Development and evaluation of an artificial intelligence system for COVID-19 diagnosis," vol. 11, no. 1, pp. 1-14, 2020.
[32]H. S. Maghdid, A. T. Asaad, K. Z. Ghafoor, A. S. Sadiq, S. Mirjalili, and M. K. Khan, "Diagnosing COVID-19 pneumonia from X-ray and CT images using deep learning and transfer learning algorithms," in Multimodal Image Exploitation and Learning 2021, 2021, vol. 11734, p. 117340E: International Society for Optics and Photonics.
[33]C. Zheng, X. Deng, Q. Fu, Q. Zhou, J. Feng, H. Ma, W. Liu, and X. J. M. Wang, "Deep learning-based detection for COVID-19 from chest CT using weak label," 2020.
[34]T. Yan, P. K. Wong, H. Ren, H. Wang, J. Wang, Y. J. C. Li, Solitons, and Fractals, "Automatic distinction between covid-19 and common pneumonia using multi-scale convolutional neural network on chest ct scans," vol. 140, p. 110153, 2020.
[35]C. Ouchicha, O. Ammor, M. J. C. Meknassi, Solitons, and Fractals, "CVDNet: A novel deep learning architecture for detection of coronavirus (Covid-19) from chest x-ray images," vol. 140, p. 110245, 2020.
[36]P. K. Sethy and S. K. Behera, "Detection of coronavirus disease (covid-19) based on deep features," 2020.
[37]F. Rustam, A. A. Reshi, A. Mehmood, S. Ullah, B.-W. On, W. Aslam, and G. S. J. I. a. Choi, "COVID-19 future forecasting using supervised machine learning models," vol. 8, pp. 101489-101499, 2020.
[38]N. S. Punn, S. K. Sonbhadra, and S. J. M. Agarwal, "COVID-19 epidemic analysis using machine learning and deep learning algorithms," 2020.
[39]M. H. D. M. Ribeiro, R. G. da Silva, V. C. Mariani, L. J. C. dos Santos Coelho, Solitons, and Fractals, "Short-term forecasting COVID-19 cumulative confirmed cases: Perspectives for Brazil," vol. 135, p. 109853, 2020.
[40]R. G. da Silva, M. H. D. M. Ribeiro, V. C. Mariani, L. J. C. dos Santos Coelho, Solitons, and Fractals, "Forecasting Brazilian and American COVID-19 cases based on artificial intelligence coupled with climatic exogenous variables," vol. 139, p. 110027, 2020.
[41]B. J. C. Ghanbari, Solitons and Fractals, "On forecasting the spread of the COVID-19 in Iran: The second wave," vol. 140, p. 110176, 2020.
[42]Z. Malki, E.-S. Atlam, A. E. Hassanien, G. Dagnew, M. A. Elhosseini, I. J. C. Gad, Solitons, and Fractals, "Association between weather data and COVID-19 pandemic predicting mortality rate: Machine learning approaches," vol. 138, p. 110137, 2020.
[43]M. E. Chowdhury, T. Rahman, A. Khandakar, R. Mazhar, M. A. Kadir, Z. B. Mahbub, K. R. Islam, M. S. Khan, A. Iqbal, and N. J. I. A. Al Emadi, "Can AI help in screening viral and COVID-19 pneumonia?," vol. 8, pp. 132665-132676, 2020.
[44]K. J. N. T. Fukushima and I. Its Applications, "Recent advances in the deep CNN neocognitron," vol. 10, no. 4, pp. 304-321, 2019.
[45]C. Nwankpa, W. Ijomah, A. Gachagan, and S. Marshall, "Activation functions: Comparison of trends in practice and research for deep learning," arXiv preprint arXiv:1811.03378, 2018.
[46]A. F. Agarap, "Deep learning using rectified linear units (relu)," arXiv preprint arXiv:1803.08375, 2018.
[47]王声柱, "基于深度学习和半监督聚类的入侵防御技术研究," 江苏科技大学, 2016.
[48]S. Ioffe and C. Szegedy, "Batch normalization: Accelerating deep network training by reducing internal covariate shift," in International conference on machine learning, 2015, pp. 448-456: PMLR.
[49]A. K. Das, S. Kalam, C. Kumar, D. J. C. Sinha, Solitons, and Fractals, "TLCoV-An automated Covid-19 screening model using Transfer Learning from chest X-ray images," vol. 144, p. 110713, 2021.
[50]D. Ardila, A. P. Kiraly, S. Bharadwaj, B. Choi, J. J. Reicher, L. Peng, D. Tse, M. Etemadi, W. Ye, and G. J. N. m. Corrado, "End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography," vol. 25, no. 6, pp. 954-961, 2019.
[51]M. L. Holshue, C. DeBolt, S. Lindquist, K. H. Lofy, J. Wiesman, H. Bruce, C. Spitters, K. Ericson, S. Wilkerson, and A. J. N. E. J. o. M. Tural, "First case of 2019 novel coronavirus in the United States," 2020.
[52]C. Wang, P. W. Horby, F. G. Hayden, and G. F. J. T. l. Gao, "A novel coronavirus outbreak of global health concern," vol. 395, no. 10223, pp. 470-473, 2020.
[53]D. Lv, Y. Wang, S. Wang, Q. Zhang, W. Qi, Y. Li, and L. J. M. P. Sun, "A Cascade‐SEME network for COVID‐19 detection in chest x‐ray images," vol. 48, no. 5, pp. 2337-2353, 2021.
[54]D. Das, K. Santosh, U. J. P. Pal, and e. s. i. medicine, "Truncated inception net: COVID-19 outbreak screening using chest X-rays," vol. 43, no. 3, pp. 915-925, 2020.
[55]Z. Han, B. Wei, Y. Hong, T. Li, J. Cong, X. Zhu, H. Wei, and W. Zhang, "Accurate screening of COVID-19 using attention-based deep 3D multiple instance learning," IEEE transactions on medical imaging, vol. 39, no. 8, pp. 2584-2594, 2020.
[56]C. Zheng, X. Deng, Q. Fu, Q. Zhou, J. Feng, H. Ma, W. Liu, and X. Wang, "Deep learning-based detection for COVID-19 from chest CT using weak label," MedRxiv, 2020.
[57]E. Hussain, M. Hasan, M. A. Rahman, I. Lee, T. Tamanna, and M. Z. Parvez, "CoroDet: A deep learning based classification for COVID-19 detection using chest X-ray images," Chaos, Solitons & Fractals, vol. 142, p. 110495, 2021.
[58]E. E.-D. Hemdan, M. A. Shouman, and M. E. Karar, "Covidx-net: A framework of deep learning classifiers to diagnose covid-19 in x-ray images," arXiv preprint arXiv:2003.11055, 2020.
[59]I. D. Apostolopoulos and T. A. Mpesiana, "Covid-19: automatic detection from x-ray images utilizing transfer learning with convolutional neural networks," Physical and Engineering Sciences in Medicine, vol. 43, no. 2, pp. 635-640, 2020.
[60]L. Wang, Z. Q. Lin, and A. Wong, "Covid-net: A tailored deep convolutional neural network design for detection of covid-19 cases from chest x-ray images," Scientific Reports, vol. 10, no. 1, pp. 1-12, 2020.
[61]P. K. Sethy, S. K. Behera, P. K. Ratha, and P. Biswas, "Detection of coronavirus disease (COVID-19) based on deep features and support vector machine," 2020.
[62]S. Wang, B. Kang, J. Ma, X. Zeng, M. Xiao, J. Guo, M. Cai, J. Yang, Y. Li, and X. Meng, "A deep learning algorithm using CT images to screen for Corona Virus Disease (COVID-19)," European Radiology, pp. 1-9, 2021.
[63]H. Panwar, P. Gupta, M. K. Siddiqui, R. Morales-Menendez, and V. Singh, "Application of deep learning for fast detection of COVID-19 in X-Rays using nCOVnet," Chaos, Solitons & Fractals, vol. 138, p. 109944, 2020.