近年來生物醫學影像已被大量數位化，人們嘗試發展演算法去分析影像，用以解決現有的問題，例如：進行乳癌偵測時，病理醫生需花費大量的時間閱讀病理切片影像，找出惡性腫瘤細胞區域；生物實驗的過程中，為了避免使用到發炎的細胞，研究員需要經過繁瑣的實驗才能驗證細胞並無發炎。過往的文獻進行影像分析時，多需要高度生物專業背景知識才能找出影像中之特徵。本研究憑藉卷積神經網路(Convolutional Neural Networks, CNNs)自行學習影像特徵之性質，用於乳癌細胞及發炎細胞影像之偵測。
本論文將分成兩部分，分別使用深度學習模型─U-Net及VGG，對乳腺癌淋巴結轉移細胞螢光影像以及MC3T3-E1小鼠骨細胞之發炎影像進行分析。乳癌細胞影像的部份，本研究使用兩種不同的分析策略對惡性腫瘤細胞進行辨識，並比較不同策略下腫瘤偵測率之差異，建立一個高準確度的影像辨識系統，迅速找出影像中腫瘤細胞的區域，其腫瘤偵測率分別為80.83%和92.04%。發炎細胞影像的部分，則針對不同時間點進行發炎實驗之發炎細胞影像進行分類，並以細胞之IL-6基因表現量作為發炎程度的依據，驗證本研究之系統的可靠性，最終我們將模型判斷之發炎特徵以可視化方式呈現，來解釋模型高準確率背後的原因。

In recent years, large quantities of biomedical images have been digitized, and people try to analyze these images with different algorithms to solve various problems. However, most of these analysis methods rely on professional biomedical knowledge. With the advance of deep learning, more and more people are trying to use Convolutional Neural Networks (CNNs) to analyze these images due to their superior ability to learn features from images directly. This study is divided into two parts. The first part is to detect malignant tumor cells in lymph node metastases from breast cancer in the images by a deep learning model. In this regard, we use two different analysis strategies to detect malignant tumor cells and discuss the difference in tumor detection rates under different methods. The tumor detection rates of the proposed models are 80.83% and 92.04%, respectively. The second part of this study is to identify whether the cells in the acquired are inflamed or not. We use the CNN model to classify inflamed cells' images and use the gene expression of IL-6 for verification. Furthermore, we explore the features extracted by our model and visualize them to explain the model's decisions.

摘要 i
Abstract ii
誌謝 iii
目錄 iv
圖目錄 vii
表目錄 viii
第一章 緒論 1
1.1 研究動機 1
1.2 文獻回顧 1
1.3 研究架構 3
第二章 深度學習 4
2.1 卷積神經網路 4
2.1.1 卷積層 4
2.1.2 池化層 5
2.1.3 全連接層 5
2.1.4 全局平均池化層 5
2.1.5 激勵函數 6
2.2 遷移學習 7
第三章 研究數據 8
3.1 乳腺癌之淋巴結轉移細胞螢光影像 8
3.1.1 數據標記 8
3.1.2 數據前處理 9
3.2 MC3T3-E1細胞影像 10
3.2.1 基因表現量 11
3.2.2 數據前處理 12
第四章 研究方法 13
4.1 基於U-Net之乳癌細胞切割 13
4.1.1 U-Net模型架構 13
4.1.2 實驗設計 14
4.2 基於VGG模型之發炎細胞偵測 16
4.2.1 VGG模型架構 16
4.2.2 特徵可視化 17
4.2.3 實驗設計 18
第五章 研究結果 19
5.1 評估指標 19
5.2 乳癌細胞切割 20
5.2.1 模型效能 20
5.2.2 結果呈現 22
5.2.3 結果討論 24
5.3 發炎細胞偵測 25
5.3.1 模型效能 25
5.3.2 發炎特徵可視化 26
5.3.3 結果討論 28
第六章 總結 29
參考資料 30

[1] N. Stathonikos, M. Veta, A. Huisman, and P. J. van Diest, "Going fully digital: Perspective of a Dutch academic pathology lab," Journal of pathology informatics, vol. 4, 2013.
[2] U. Djuric, G. Zadeh, K. Aldape, and P. Diamandis, "Precision histology: how deep learning is poised to revitalize histomorphology for personalized cancer care," NPJ precision oncology, vol. 1, no. 1, pp. 1-5, 2017.
[3] S. S. Raab et al., "Clinical impact and frequency of anatomic pathology errors in cancer diagnoses," Cancer: Interdisciplinary International Journal of the American Cancer Society, vol. 104, no. 10, pp. 2205-2213, 2005.
[4] R. E. Nakhleh, "Error reduction in surgical pathology," Archives of pathology & laboratory medicine, vol. 130, no. 5, pp. 630-632, 2006.
[5] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," nature, vol. 521, no. 7553, pp. 436-444, 2015.
[6] G. Litjens et al., "Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis," Scientific reports, vol. 6, p. 26286, 2016.
[7] C. Cortes and V. Vapnik, "Support-vector networks," Machine learning, vol. 20, no. 3, pp. 273-297, 1995.
[8] J. Friedman, T. Hastie, and R. Tibshirani, "Additive logistic regression: a statistical view of boosting (with discussion and a rejoinder by the authors)," The annals of statistics, vol. 28, no. 2, pp. 337-407, 2000.
[9] Y. Rejani and S. T. Selvi, "Early detection of breast cancer using SVM classifier technique," arXiv preprint arXiv:0912.2314, 2009.
[10] Y.-L. Huang, K.-L. Wang, and D.-R. Chen, "Diagnosis of breast tumors with ultrasonic texture analysis using support vector machines," Neural Computing & Applications, vol. 15, no. 2, pp. 164-169, 2006.
[11] M. Mohammadpoor, A. Shoeibi, H. Zare, and H. Shojaee, "A hierarchical classification method for breast tumor detection," Iranian Journal of Medical Physics, vol. 13, no. 4, pp. 261-268, 2016.
[12] 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.
[13] O. Russakovsky et al., "Imagenet large scale visual recognition challenge," International journal of computer vision, vol. 115, no. 3, pp. 211-252, 2015.
[14] K. Simonyan and A. Zisserman, "Very deep convolutional networks for large-scale image recognition," arXiv preprint arXiv:1409.1556, 2014.
[15] C. Szegedy et al., "Going deeper with convolutions," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2015, pp. 1-9.
[16] 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.
[17] G. Litjens et al., "A survey on deep learning in medical image analysis," Medical image analysis, vol. 42, pp. 60-88, 2017.
[18] D. Shen, G. Wu, and H.-I. Suk, "Deep learning in medical image analysis," Annual review of biomedical engineering, vol. 19, pp. 221-248, 2017.
[19] D. Wang, A. Khosla, R. Gargeya, H. Irshad, and A. H. Beck, "Deep learning for identifying metastatic breast cancer," arXiv preprint arXiv:1606.05718, 2016.
[20] A. Cruz-Roa et al., "Accurate and reproducible invasive breast cancer detection in whole-slide images: A Deep Learning approach for quantifying tumor extent," Scientific reports, vol. 7, p. 46450, 2017.
[21] Z. Han, B. Wei, Y. Zheng, Y. Yin, K. Li, and S. Li, "Breast cancer multi-classification from histopathological images with structured deep learning model," Scientific reports, vol. 7, no. 1, pp. 1-10, 2017.
[22] T. Q. Y.-W. Tsang, D. Epstein, and N. Rajpoot, "Tumor Segmentation in Whole Slide Histology Images using Persistent Homology and Deep Convolutional Features."
[23] B. Gecer, S. Aksoy, E. Mercan, L. G. Shapiro, D. L. Weaver, and J. G. Elmore, "Detection and classification of cancer in whole slide breast histopathology images using deep convolutional networks," Pattern recognition, vol. 84, pp. 345-356, 2018.
[24] Z. Qin, F. Yu, C. Liu, and X. Chen, "How convolutional neural network see the world-A survey of convolutional neural network visualization methods," arXiv preprint arXiv:1804.11191, 2018.
[25] D. Erhan, Y. Bengio, A. Courville, and P. Vincent, "Visualizing higher-layer features of a deep network," University of Montreal, vol. 1341, no. 3, p. 1, 2009.
[26] A. Nguyen, A. Dosovitskiy, J. Yosinski, T. Brox, and J. Clune, "Synthesizing the preferred inputs for neurons in neural networks via deep generator networks," in Advances in neural information processing systems, 2016, pp. 3387-3395.
[27] J. Yosinski, J. Clune, A. Nguyen, T. Fuchs, and H. Lipson, "Understanding neural networks through deep visualization," arXiv preprint arXiv:1506.06579, 2015.
[28] M. D. Zeiler, G. W. Taylor, and R. Fergus, "Adaptive deconvolutional networks for mid and high level feature learning," in 2011 International Conference on Computer Vision, 2011: IEEE, pp. 2018-2025.
[29] M. D. Zeiler and R. Fergus, "Visualizing and understanding convolutional networks," in European conference on computer vision, 2014: Springer, pp. 818-833.
[30] K. Simonyan, A. Vedaldi, and A. Zisserman, "Deep inside convolutional networks: Visualising image classification models and saliency maps," 2014.
[31] J. T. Springenberg, A. Dosovitskiy, T. Brox, and M. Riedmiller, "Striving for simplicity: The all convolutional net," arXiv preprint arXiv:1412.6806, 2014.
[32] B. Zhou, A. Khosla, A. Lapedriza, A. Oliva, and A. Torralba, "Learning deep features for discriminative localization," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 2921-2929.
[33] R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, "Grad-cam: Visual explanations from deep networks via gradient-based localization," in Proceedings of the IEEE international conference on computer vision, 2017, pp. 618-626.
[34] A. Chattopadhay, A. Sarkar, P. Howlader, and V. N. Balasubramanian, "Grad-cam++: Generalized gradient-based visual explanations for deep convolutional networks," in 2018 IEEE Winter Conference on Applications of Computer Vision (WACV), 2018: IEEE, pp. 839-847.
[35] Y. LeCun, L. Bottou, Y. Bengio, and P. Haffner, "Gradient-based learning applied to document recognition," Proceedings of the IEEE, vol. 86, no. 11, pp. 2278-2324, 1998.
[36] M. Lin, Q. Chen, and S. Yan, "Network in network," arXiv preprint arXiv:1312.4400, 2013.
[37] B. Xu, N. Wang, T. Chen, and M. Li, "Empirical evaluation of rectified activations in convolutional network," arXiv preprint arXiv:1505.00853, 2015.
[38] S. J. Pan and Q. Yang, "A survey on transfer learning," IEEE Transactions on knowledge and data engineering, vol. 22, no. 10, pp. 1345-1359, 2009.
[39] H.-C. Shin et al., "Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning," IEEE transactions on medical imaging, vol. 35, no. 5, pp. 1285-1298, 2016.
[40] N. Tajbakhsh et al., "Convolutional neural networks for medical image analysis: Full training or fine tuning?," IEEE transactions on medical imaging, vol. 35, no. 5, pp. 1299-1312, 2016.
[41] S. Kornblith, J. Shlens, and Q. V. Le, "Do better imagenet models transfer better?," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2019, pp. 2661-2671.
[42] J. Yosinski, J. Clune, Y. Bengio, and H. Lipson, "How transferable are features in deep neural networks?," in Advances in neural information processing systems, 2014, pp. 3320-3328.
[43] P. Bankhead et al., "QuPath: Open source software for digital pathology image analysis," Scientific reports, vol. 7, no. 1, pp. 1-7, 2017.
[44] D. Komura and S. Ishikawa, "Machine learning methods for histopathological image analysis," Computational and structural biotechnology journal, vol. 16, pp. 34-42, 2018.
[45] S. Grahama, M. Shabana, T. Qaisera, N. A. Koohbanania, and S. Ali, "Classification of Lung Cancer Histology Images using Patch-Level Summary Statistics."
[46] N. Coudray et al., "Classification and mutation prediction from non–small cell lung cancer histopathology images using deep learning," Nature medicine, vol. 24, no. 10, pp. 1559-1567, 2018.
[47] S. C. Wong, A. Gatt, V. Stamatescu, and M. D. McDonnell, "Understanding data augmentation for classification: when to warp?," in 2016 international conference on digital image computing: techniques and applications (DICTA), 2016: IEEE, pp. 1-6.
[48] C. Shorten and T. M. Khoshgoftaar, "A survey on image data augmentation for deep learning," Journal of Big Data, vol. 6, no. 1, p. 60, 2019.
[49] L. Overbergh, A. Giulietti, D. Valckx, B. Decallonne, R. Bouillon, and C. Mathieu, "The use of real-time reverse transcriptase PCR for the quantification of cytokine gene expression," Journal of biomolecular techniques: JBT, vol. 14, no. 1, p. 33, 2003.
[50] S. Dhanasekaran, T. M. Doherty, J. Kenneth, and T. T. S. Group, "Comparison of different standards for real-time PCR-based absolute quantification," Journal of immunological methods, vol. 354, no. 1-2, pp. 34-39, 2010.
[51] O. Ronneberger, P. Fischer, and T. Brox, "U-net: Convolutional networks for biomedical image segmentation," in International Conference on Medical image computing and computer-assisted intervention, 2015: Springer, pp. 234-241.
[52] J. Long, E. Shelhamer, and T. Darrell, "Fully convolutional networks for semantic segmentation," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2015, pp. 3431-3440.
[53] N. Srivastava, G. Hinton, A. Krizhevsky, I. Sutskever, and R. Salakhutdinov, "Dropout: a simple way to prevent neural networks from overfitting," The journal of machine learning research, vol. 15, no. 1, pp. 1929-1958, 2014.
[54] G. Klambauer, T. Unterthiner, A. Mayr, and S. Hochreiter, "Self-normalizing neural networks," in Advances in neural information processing systems, 2017, pp. 971-980.
[55] J. A. Hanley and B. J. McNeil, "The meaning and use of the area under a receiver operating characteristic (ROC) curve," Radiology, vol. 143, no. 1, pp. 29-36, 1982.