研究生: |
劉員宏 Yuan-Hung Liu |
---|---|
論文名稱: |
基於資料擴增、不確定性標籤處理與集成學習之胸腔疾病分類 Thorax Symptom Classification Based on Data Augmentation, Uncertainty Label Handling, and Multi-model Ensemble |
指導教授: |
林昌鴻
Chang-Hong Lin |
口試委員: |
林昌鴻
Chang-Hong Lin 陳永耀 林敬舜 Ching-Shun Lin 王煥宗 Huan-Chun Wang 陳維美 Wei-Mei Chen |
學位類別: |
碩士 Master |
系所名稱: |
電資學院 - 電子工程系 Department of Electronic and Computer Engineering |
論文出版年: | 2020 |
畢業學年度: | 108 |
語文別: | 英文 |
論文頁數: | 85 |
中文關鍵詞: | 醫學影像分類 、不確定性標籤處理 、深度學習 、卷積神經網路 (CNN) 、資料擴增 、集成學習 、加權損失函數 |
外文關鍵詞: | Medical image Classification, Uncertainty Label Handling, Deep Learning, Convolution Neural Networks (CNNs), Data Augmentation, Model Ensemble, Weighted loss |
相關次數: | 點閱:431 下載:0 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
進年來,深度學習技術被廣泛得應用在影像處理上。其中一個熱門的應用是醫學影像分類處理。胸腔X光攝影是一個常見且有不錯品質的胸腔疾病診斷方式。然而,訓練一個合格的放射科醫師需要很長的時間,而且,即便是專業的放射科醫師依然會有診斷錯誤的時候。因此,人們有了使用電腦輔助診斷的想法。因為在胸腔X光影像的資料集當中,胸腔在照片中佔有的比例、身體的旋轉及位移是很常見的,另外,不確定標籤也是另一種醫學資料集中常見的問題。針對這些問題,我們在此篇論文中提出了一些方法。針對不確定標籤,我們使用了不確定性標籤處理方法。為了增加資料的多樣性及模擬一些胸腔X光影像資料集中常見的情況,我們提出了一些資料擴增的方法。另外,為了進一步提升,我們使用了集成學習及加權損失函數。在我們的實驗中,我們使用CheXpert 資料集,在三種流行的卷積網路架構平均上,我們的最好模型獲得了0.880 的AUC 5 ,並相較於沒有使用任何方法的模型提高了0.030。
The deep learning technique has been widely used for image processing recently. One of the most popular subjects is medical image classification. The chest radiograph is the most common way to diagnose thorax disease and had good success. However, training a qualified radiologist needs a long time and the diagnosis still exists mistakes. Computer-aided diagnosis, therefore, has been proposed. Since we observed that within existing chest X-ray image datasets, it is prevalent that images have different chest occupancy ratios, and the bodies might be biased or rotated. Moreover, the uncertain labels are also common in the datasets. In this thesis, we proposed some methods to alleviate these problems. The uncertainty labels handling strategies are used to handle the uncertainty labels, and the data augmentation strategies are used to increase the data diversity and simulate different chest occupancy ratios, rotated and biased bodies in the chest X-ray datasets. For further improvement, we applied model ensemble and weighted loss to improved the accuracy of diagnosis. In the experiment, on the CheXpert dataset, by using the proposed methods, on average of three popular architectures, the best model score an AUC 5 of 0.880, which is 0.030 better than the model without any modification.
[1] F. Chollet, "Xception: Deep learning with depthwise separable convolutions," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 1251-1258.
[2] G. Huang, Z. Liu, L. van der Maaten, and K.Q. Weinberger, "Densely connected convolutional networks," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 4700-4708.
[3] K. Fukushima, and S. Miyake, "Neocognitron: A self-organizing neural network model for a mechanism of visual pattern recognition," in Competition and Cooperation in Neural Nets: Springer, 1982, pp. 267-285.
[4] A. Krizhevsky, I. Sutskever, and G.E. Hinton, "Imagenet classification with deep convolutional neural networks," in Advances in Neural Information Processing Systems, 2012, pp. 1097-1105.
[5] K. Simonyan, and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," in International Conference on Learning Representations, 2015.
[6] Y. LeCun, B. Boser, J.S. Denker, and D. Henderson, "Backpropagation applied to handwritten zip code recognition," Neural Computation, 1989, vol. 1, no. 4, pp. 541-551.
[7] R. Ranjan, S. Sankaranarayanan, C.D. Castillo, and R. Chellappa, "An all-in-one convolutional neural network for face analysis," in 2017 12th IEEE International Conference on Automatic Face & Gesture Recognition, 2017, pp. 17-24.
[8] R. Laroca, E. Severo, L.A. Zanlorensi, L.S. Oliveira, G.R. Gonçalves, W.R. Schwartz, and D. Menotti, "A robust real-time automatic license plate recognition based on the YOLO detector," in 2018 International Joint Conference on Neural Networks, 2018, pp. 1-10.
[9] A. R. Lopez, X. Giro-i-Nieto, J. Burdick, and O. Marques, "Skin lesion classification from dermoscopic images using deep learning techniques," in 2017 13th IASTED International Conference on Biomedical Engineering, 2017, pp. 49-54.
[10] Y. Jiang, R.M. Nishikawa, R.A. Schmidt, C.E. Metz, M.L. Giger, and K. Doi, "Improving breast cancer diagnosis with computer-aided diagnosis," Academic Radiology, 1999, vol. 6, no. 1, pp. 22-33.
[11] J. Tang, R.M. Rangayyan, J. Xu, I. El Naqa, and Y. Yang, "Computer-aided detection and diagnosis of breast cancer with mammography: recent advances," IEEE Transactions on Information Technology in Biomedicine, 2009, vol. 13, no. 2, pp. 236-251.
[12] P. Huang, S. Park, R. Yan, J. Lee, L.C. Chu, C.T. Lin, A. Hussien, J. Rathmell, B. Thomas, C. Chen, R. Hales, D.S. Ettinger , M. Brock, P. Hu, E.K. Fishman, E. Gabrielson, and S. Lam, "Added value of computer-aided CT image features for early lung cancer diagnosis with small pulmonary nodules: a matched case-control study," Radiology, 2018, vol. 286, no. 1, pp. 286-295.
[13] A. Esteva, B. Kuprel, R.A. Novoa, J. Ko, S.M. Swetter, H.M. Blau, and S. Thrun, "Dermatologist-level classification of skin cancer with deep neural networks," Nature, 2017, vol. 542, no. 7639, pp. 115-118.
[14] S. Ryoo, and H.J. Kim, "Activities of the Korean institute of tuberculosis," Osong Public Health and Research Perspectives, 2014, vol. 5, pp. S43-S49.
[15] A.E.W. Johnson, T.J. Pollard, N.R. Greenbaum, M.P. Lungren, C.Y Deng, Y.F. Peng, Z.Y. Lu, R.G. Mark, S.J. Berkowitz, and S. Horng, "MIMIC-CXR-JPG, a large publicly available database of labeled chest radiographs," 2019, arXiv preprint arXiv:1901.07042.
[16] Y. Peng, X. Wang, L. Lu, M. Bagheri, R. Summers, and Z. Lu, "Negbio: a high-performance tool for negation and uncertainty detection in radiology reports," American Medical Informatics Association Summits on Translational Science Proceedings, 2018, vol. 2018, pp. 188-196.
[17] J. Irvin, P. Rajpurkar, M. Ko, Y.F. Yu, S. Ciurea-Ilcus, C. Chute, H. Marklund, B. Haghgoo, R. Ball, K. Shpanskaya, J. Seekins, D.A. Mong, S.S. Halabi, J.K. Sandberg, R. Jones, D.B. Larson, C.P. Langlotz, B.N. Patel, M.P. Lungren, and A.Y. Ng, "Chexpert: A large chest radiograph dataset with uncertainty labels and expert comparison," in Proceedings of the Association for the Advancement of Artificial Intelligence Conference on Artificial Intelligence, 2019, vol. 33, pp. 590-597.
[18] C. Szegedy, S. Ioffe, V. Vanhoucke, and A. A. Alemi, "Inception-v4, inception-resnet and the impact of residual connections on learning," in Thirty-first Association for the Advancement of Artificial Intelligence Conference on Artificial Intelligence, 2017.
[19] C. Szegedy, W. Liu, Y.Q. Jia, P. Sermanet, S. Reed, D. Anguelov, D. Erhan, V. Vanhoucke, and A. Rabinovich, "Going deeper with convolutions," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015, pp. 1-9.
[20] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, "Rethinking the inception architecture for computer vision," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016, pp. 2818-2826.
[21] 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.
[22] A.G. Howard, M.L. Zhu, B. Chen, D. Kalenichenko, W.J. Wang, T. Weyand, M. Andreetto, and H. Adam, "Mobilenets: Efficient convolutional neural networks for mobile vision applications," 2017, arXiv preprint arXiv:1704.04861.
[23] J.A. Hanley, and B.J. McNeil, "The meaning and use of the area under a receiver operating characteristic (ROC) curve," in Radiology, 1982, vol. 143, no. 1, pp. 29-36.
[24] R. Müller, S. Kornblith, and G.E. Hinton, "When does label smoothing help?," in Advances in Neural Information Processing Systems, 2019, pp. 4694-4703.
[25] X. Wang, Y. Peng, L. Lu, Z. Lu, M. Bagheri, and R.M. Summers, "Chestx-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 2097-2106.
[26] D. Opitz and R. Maclin, "Popular ensemble methods: An empirical study," in Journal of Artificial Intelligence Research, 1999, vol. 11, pp. 169-198.
[27] P. Rajpurkar, J. Irvin, K.L. Zhu, B. Yang, H. Mehta, T. Duan, D. Ding, A. Bagul, C. Langlotz, K. Shpanskaya, M.P. Lungren, and A.Y. Ng, "Chexnet: Radiologist-level pneumonia detection on chest x-rays with deep learning," 2017, arXiv preprint arXiv:1711.05225.
[28] D.P. Kingma and J. Ba, "Adam: A method for stochastic optimization," 2014, arXiv preprint arXiv:1412.6980.
[29] "Python Website." [Online]. [https://www.python.org/] (accessed July, 2020).
[30] "Keras Website." [Online]. [https://keras.io/] (accessed July,2020).
[31] "TensorFlow Website." [Online]. [https://www.tensorflow.org/] (accessed July,2020).