乳腺癌是女性癌症死亡的第二大原因。儘管癌症治療方法之有所進展，轉移性乳腺癌仍然會導致高死亡率，準確預測和早期檢測乳腺癌轉移狀態對乳腺癌患者的預後是相當重要的。腋窩淋巴節轉移之狀態是評估癌症分期和決定乳腺癌患者治療策略的重要指標。因此，我們開發了一種基於卷積神經網絡之新型電腦輔助預測系統，該系統使用乳房超音波影像來預測腋窩淋巴節之狀況。首先，採用Mask R-CNN模型從整張乳房超音波偵測出腫瘤之位置並切割出腫瘤區域。在獲得腫瘤區域後，我們提取腫瘤周圍區域並使用DenseNet模型進行腋窩淋巴節狀況之預測。在我們的實驗中，共有153例乳腺腫瘤患者, 其中有59例腋窩淋巴節有轉移，94例腋窩淋巴節沒有轉移，用於評估我們提出的方法。根據我們實驗的結果，最佳預測表現是使用包含腫瘤和腫瘤周圍3公厘之區域。其準確性，靈敏度和特異性分別為81.05％（124/153），81.36％（48/59）和80.85％（76/94），ROC曲線下的面積為0.8054。我們所提出的電腦輔助預測系統，結合原發腫瘤的腫瘤和腫瘤周圍區域，在乳癌患者中，可以有效的預測腋窩淋巴節轉移的狀況。

Breast cancer is the second leading cause of cancer death in women. Despite the improvement in cancer therapy, metastatic breast cancer still leads to a high mortality rate. Accurate prediction and early detection of breast cancer metastasis status are important for the prognosis of breast cancer patients. Axillary lymph nodes (ALN) status is an important indicator in assessing cancer staging and deciding the treatment strategy for patients with breast cancer. Consequently, we developed a novel computer-aided prediction (CAP) system based on convolutional neural networks (CNN) using breast ultrasound to distinguish the ALN status. At first, the Mask R-CNN model was used to detect the tumor location and segment the tumor region from the whole US image. After obtained the tumor region, we extracted the peritumoral region and used the DenseNet model to predict the ALN status. In the experiments, 153 patients with breast tumor composed of 59 cases with ALN metastasis and 94 cases without ALN metastasis, used to evaluate our proposed method. According to results, the best prediction performance was using tumor region with the peritumoral region (3mm), the accuracy, sensitivity, specificity, and AUC are 81.05% (124/153), 81.36% (48/59) and 80.85% (76/94), and 0.8054. In summary, the proposed CAP model combines the primary tumor and peritumoral region, which can be an effective method to prediction the ALN status in patients with breast cancer.

口試委員會審定書 i
致謝 ii
摘要 iii
Abstract iv
Table of Contents vi
List of Figures viii
List of Tables x
Chapter 1 Introduction 1
Chapter 2 Material 4
Chapter 3 Methods 6
3.1. Tumor detection and segmentation 7
3.1.1. Mask R-CNN 7
3.1.2. Training strategies 11
3.1.3. Tumor region extraction 14
3.2. Peritumoral region extraction 14
3.3. ALN status prediction 16
3.3.1. CNN model 16
3.3.2. Machine learning models 18
Chapter 4 Experiment Results 20
4.1. Experimental environment 20
4.2. Statistical Analysis 20
4.3. Results 21
4.3.1. Comparison of different input region 21
4.3.2. Comparison of the proposed method with ML models 28
Chapter 5 Conclusions and Discussion 32
References 36

[1] R. L. Siegel, K. D. Miller, and A. Jemal, "Cancer statistics, 2019," CA: a cancer journal for clinicians, vol. 69, no. 1, pp. 7-34, Jan 2019.
[2] C. E. DeSantis, J. Ma, A. Goding Sauer, L. A. Newman, and A. Jemal, "Breast cancer statistics, 2017, racial disparity in mortality by state," CA: a cancer journal for clinicians, vol. 67, no. 6, pp. 439-448, Oct 2017.
[3] A. Alitalo and M. Detmar, "Interaction of tumor cells and lymphatic vessels in cancer progression," Oncogene, vol. 31, no. 42, p. 4499, Dec 2012.
[4] U. Veronesi, F. Rilke, A. Luini, V. Sacchini, V. Galimberti, T. Campa, E. D. Bei, M. Greco, A. Magni, and M. Merson, "Distribution of axillary node metastases by level of invasion. An analysis of 539 cases," Cancer, vol. 59, no. 4, pp. 682-687, Feb 1987.
[5] U. Veronesi, A. Luini, V. Galimberti, S. Marchini, V. Sacchini, and F. Rilke, "Extent of metastatic axillary involvement in 1446 cases of breast cancer," European journal of surgical oncology: the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology, vol. 16, no. 2, pp. 127-133, Apr 1990.
[6] C. L. Carter, C. Allen, and D. E. Henson, "Relation of tumor size, lymph node status, and survival in 24,740 breast cancer cases," Cancer, vol. 63, no. 1, pp. 181-187, Jan 1989.
[7] B. Weigelt, J. L. Peterse, and L. J. Van''t Veer, "Breast cancer metastasis: markers and models," Nature reviews cancer, vol. 5, no. 8, p. 591, Aug 2005.
[8] D. Ivens, A. Hoe, T. Podd, C. Hamilton, I. Taylor, and G. Royle, "Assessment of morbidity from complete axillary dissection," British journal of cancer, vol. 66, no. 1, p. 136, Jul 1992.
[9] S.-Q. Qiu, H.-C. Zeng, F. Zhang, C. Chen, W.-H. Huang, R. G. Pleijhuis, J.-D. Wu, G. M. Van Dam, and G.-J. Zhang, "A nomogram to predict the probability of axillary lymph node metastasis in early breast cancer patients with positive axillary ultrasound," Scientific reports, vol. 6, p. 21196, Feb 2016.
[10] S. Koscielny, M. Tubiana, M. Le, A. Valleron, H. Mouriesse, G. Contesso, and D. Sarrazin, "Breast cancer: relationship between the size of the primary tumour and the probability of metastatic dissemination," British journal of cancer, vol. 49, no. 6, p. 709, Jun 1984.
[11] M. Takada, M. Sugimoto, Y. Naito, H.-G. Moon, W. Han, D.-Y. Noh, M. Kondo, K. Kuroi, H. Sasano, and T. Inamoto, "Prediction of axillary lymph node metastasis in primary breast cancer patients using a decision tree-based model," BMC medical informatics and decision making, vol. 12, no. 1, p. 54, Jun 2012.
[12] W. K. Moon, Y.-W. Lee, Y.-S. Huang, S. H. Lee, M. S. Bae, A. Yi, C.-S. Huang, and R.-F. Chang, "Computer-aided prediction of axillary lymph node status in breast cancer using tumor surrounding tissue features in ultrasound images," Computer methods and programs in biomedicine, vol. 146, pp. 143-150, Jul 2017.
[13] W. K. Moon, I.-L. Chen, A. Yi, M. S. Bae, S. U. Shin, and R.-F. Chang, "Computer-aided prediction model for axillary lymph node metastasis in breast cancer using tumor morphological and textural features on ultrasound," Computer methods and programs in biomedicine, vol. 162, pp. 129-137, Aug 2018.
[14] X. Cui, N. Wang, Y. Zhao, S. Chen, S. Li, M. Xu, and R. Chai, "Preoperative Prediction of Axillary Lymph Node Metastasis in Breast Cancer using Radiomics Features of DCE-MRI," Scientific reports, vol. 9, no. 1, p. 2240, Feb 2019.
[15] J. Yang, T. Wang, L. Yang, Y. Wang, H. Li, X. Zhou, W. Zhao, J. Ren, X. Li, and J. Tian, "Preoperative Prediction of Axillary Lymph Node Metastasis in Breast Cancer Using Mammography-Based Radiomics Method," Scientific reports, vol. 9, no. 1, p. 4429, Mar 2019.
[16] W. Li, J. Li, K. V. Sarma, K. C. Ho, S. Shen, B. S. Knudsen, A. Gertych, and C. W. Arnold, "Path R-CNN for prostate cancer diagnosis and gleason grading of histological images," IEEE transactions on medical imaging, vol. 38, no. 4, pp. 945-954, Apr 2018.
[17] Z. Zhou, M. M. R. Siddiquee, N. Tajbakhsh, and J. Liang, "Unet++: A nested u-net architecture for medical image segmentation," in Deep Learning in Medical Image Analysis and Multimodal Learning for Clinical Decision Support, 2018, pp. 3-11.
[18] X. Qi, L. Zhang, Y. Chen, Y. Pi, Y. Chen, Q. Lv, and Z. Yi, "Automated diagnosis of breast ultrasonography images using deep neural networks," Medical image analysis, vol. 52, pp. 185-198, Feb 2019.
[19] J. G. Elmore, K. Armstrong, C. D. Lehman, and S. W. Fletcher, "Screening for breast cancer," Jama, vol. 293, no. 10, pp. 1245-1256, Mar 2005.
[20] K. He, G. Gkioxari, P. Dollár, and R. Girshick, "Mask r-cnn," in Proceedings of the IEEE international conference on computer vision, 2017, pp. 2961-2969.
[21] S. Ren, K. He, R. Girshick, and J. Sun, "Faster r-cnn: Towards real-time object detection with region proposal networks," in Advances in neural information processing systems, 2015, pp. 91-99.
[22] T.-Y. Lin, P. Dollár, R. Girshick, K. He, B. Hariharan, and S. Belongie, "Feature pyramid networks for object detection," in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2017, pp. 2117-2125.
[23] H. Chen, Q. Dou, D. Ni, J.-Z. Cheng, J. Qin, S. Li, and P.-A. Heng, "Automatic fetal ultrasound standard plane detection using knowledge transferred recurrent neural networks," in International Conference on Medical Image Computing and Computer-Assisted Intervention, 2015, pp. 507-514.
[24] H.-C. Shin, H. R. Roth, M. Gao, L. Lu, Z. Xu, I. Nogues, J. Yao, D. Mollura, and R. M. Summers, "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, May 2016.
[25] D. F. Quail and J. A. Joyce, "Microenvironmental regulation of tumor progression and metastasis," Nature medicine, vol. 19, no. 11, p. 1423, Nov 2013.
[26] H. A. Goubran, R. R. Kotb, J. Stakiw, M. E. Emara, and T. Burnouf, "Regulation of tumor growth and metastasis: the role of tumor microenvironment," Cancer growth and metastasis, vol. 7, p. CGM. S11285, Jun 2014.
[27] N. Beig, M. Khorrami, M. Alilou, P. Prasanna, N. Braman, M. Orooji, S. Rakshit, K. Bera, P. Rajiah, and J. Ginsberg, "Perinodular and intranodular radiomic features on lung CT images distinguish adenocarcinomas from granulomas," Radiology, vol. 290, no. 3, pp. 783-792, Dec 2018.
[28] 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.
[29] 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.
[30] D. W. Hosmer Jr, S. Lemeshow, and R. X. Sturdivant, Applied logistic regression. 2013.
[31] M. A. Hearst, S. T. Dumais, E. Osuna, J. Platt, and B. Scholkopf, "Support vector machines," IEEE Intelligent Systems and their applications, vol. 13, no. 4, pp. 18-28, 1998.
[32] T. Chen and C. Guestrin, "Xgboost: A scalable tree boosting system," in Proceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 2016, pp. 785-794.
[33] F. Chollet, "Keras," ed, 2015.
[34] F. Pedregosa, G. Varoquaux, A. Gramfort, V. Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, and V. Dubourg, "Scikit-learn: Machine learning in Python," Journal of machine learning research, vol. 12, pp. 2825-2830, Oct 2011.
[35] J. D. Rodriguez, A. Perez, and J. A. Lozano, "Sensitivity analysis of k-fold cross validation in prediction error estimation," IEEE transactions on pattern analysis and machine intelligence, vol. 32, no. 3, pp. 569-575, Mar 2009.
[36] D. A. Berry, K. A. Cronin, S. K. Plevritis, D. G. Fryback, L. Clarke, M. Zelen, J. S. Mandelblatt, A. Y. Yakovlev, J. D. F. Habbema, and E. J. Feuer, "Effect of screening and adjuvant therapy on mortality from breast cancer," New England Journal of Medicine, vol. 353, no. 17, pp. 1784-1792, Oct 2005.
[37] L. Tabar, A. Gad, L. Holmberg, U. Ljungquist, K. C. P. Group, C. Fagerberg, L. Baldetorp, O. Gröntoft, B. Lundström, and J. Månson, "Reduction in mortality from breast cancer after mass screening with mammography: randomised trial from the Breast Cancer Screening Working Group of the Swedish National Board of Health and Welfare," The Lancet, vol. 325, no. 8433, pp. 829-832, Apr 1985.
[38] M.-Q. Gao, B. G. Kim, S. Kang, Y. P. Choi, H. Park, K. S. Kang, and N. H. Cho, "Stromal fibroblasts from the interface zone of human breast carcinomas induce an epithelial–mesenchymal transition-like state in breast cancer cells in vitro," J Cell Sci, vol. 123, no. 20, pp. 3507-3514, Jul 2010.
[39] S. Zhang, S. Yi, D. Zhang, M. Gong, Y. Cai, and L. Zou, "Intratumoral and peritumoral lymphatic vessel density both correlate with lymph node metastasis in breast cancer," Scientific reports, vol. 7, p. 40364, Jan 2017.
[40] N. M. Braman, M. Etesami, P. Prasanna, C. Dubchuk, H. Gilmore, P. Tiwari, D. Plecha, and A. Madabhushi, "Intratumoral and peritumoral radiomics for the pretreatment prediction of pathological complete response to neoadjuvant chemotherapy based on breast DCE-MRI," Breast Cancer Research, vol. 19, no. 1, p. 57, Dec 2017.
[41] L. Zou, S. Yu, T. Meng, Z. Zhang, X. Liang, and Y. Xie, "A Technical Review of Convolutional Neural Network-Based Mammographic Breast Cancer Diagnosis," Computational and mathematical methods in medicine, vol. 2019, Mar 2019.
[42] Y. LeCun, Y. Bengio, and G. Hinton, "Deep learning," nature, vol. 521, no. 7553, p. 436, May 2015