請用此 Handle URI 來引用此文件：
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/21722完整後設資料紀錄
| DC 欄位 | 值 | 語言 |
|---|---|---|
| dc.contributor.advisor | 陳中明 | |
| dc.contributor.author | Joseph Chang | en |
| dc.contributor.author | 張漢庭 | zh_TW |
| dc.date.accessioned | 2021-06-08T03:44:01Z | - |
| dc.date.copyright | 2019-06-12 | |
| dc.date.issued | 2019 | |
| dc.date.submitted | 2019-05-10 | |
| dc.identifier.citation | [1] Bray, Freddie, et al. 'Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries.' CA: a cancer journal for clinicians 68.6 (2018): 394-424.
[2] Skandan, S. P. (2016). 5 year Overall survival of triple negative breast cancer: A single institution experience. [3] Tabár, L., Vitak, B., Chen, H. H. T., Yen, M. F., Duffy, S. W., & Smith, R. A. (2001). Beyond randomized controlled trials. Cancer, 91(9), 1724-1731. [4] Tabar, L., Yen, M. F., Vitak, B., Chen, H. H. T., Smith, R. A., & Duffy, S. W. (2003). Mammography service screening and mortality in breast cancer patients: 20-year follow-up before and after introduction of screening. The Lancet, 361(9367), 1405- 1410. [5] Nelson, H. D., Fu, R., Cantor, A., Pappas, M., Daeges, M., & Humphrey, L. (2016). Effectiveness of breast cancer screening: systematic review and meta-analysis to update the 2009 US Preventive Services Task Force recommendation. Annals of internal medicine, 164(4), 244-255. [6] Morrell, S., Taylor, R., Roder, D., Robson, B., Gregory, M., & Craig, K. (2017). Mammography service screening and breast cancer mortality in New Zealand: a National Cohort Study 1999–2011. British journal of cancer, 116(6), 828. [7] Heine, J. J., & Malhotra, P. (2002). Mammographic Tissue, Breast Cancer Risk, Serial Image Analysis, and Digital Mammography: Part 2. Serial Breast Tissue Chage and Related Temporal Influences. Academic Radiology, 9(3), 317-335. [8] Boyd, N. F., Dite, G. S., Stone, J., Gunasekara, A., English, D. R., McCredie, M. R., ... & Hopper, J. L. (2002). Heritability of mammographic density, a risk factor for breast cancer. New England Journal of Medicine, 347(12), 886-894. [9] McCormack, V. A., & dos Santos Silva, I. (2006). Breast density and parenchymal 54 patterns as markers of breast cancer risk: a meta-analysis. Cancer Epidemiology and Prevention Biomarkers, 15(6), 1159-1169. [10] Subashini, T. S., Ramalingam, V., & Palanivel, S. (2010). Automated assessment of breast tissue density in digital mammograms.Computer Vision and Image Understanding, 114(1), 33-43. [11] Stomper, P. C., Geradts, J., Edge, S. B., & Levine, E. G. (2003). Mammographic predictors of the presence and size of invasive carcinomas associated with malignant microcalcification lesions without a mass. American Journal of Roentgenology, 181(6), 1679-1684.. [12] Akselrod-Ballin, A., Karlinsky, L., Alpert, S., Hasoul, S., Ben-Ari, R., & Barkan, E. (2016). A region based convolutional network for tumor detection and classification in breast mammography. InDeep Learning and Data Labeling for Medical Applications (pp. 197-205). Springer, Cham. [13] American College of Radiology (ACR): Illustrated Breast Imaging Reporting and Data System (BI-RADS). ACR (1998) [14] Kanso, H., Hourani, R., Aoun, N., & Ghossain, M. (2009). BI-RADS: what do we need to know? Advantages and limitations. Le Journal medical libanais. The Lebanese medical journal, 57(2), 75-82. [15] Bird, R. E., Wallace, T. W., & Yankaskas, B. C. (1992). Analysis of cancers missed at screening mammography. Radiology, 184(3), 613-617. [16] Brem, R. F., Hoffmeister, J. W., Rapelyea, J. A., Zisman, G., Mohtashemi, K., Jindal, G., ... & Rogers, S. K. (2005). Impact of breast density on computer-aided detection for breast cancer. American Journal of Roentgenology, 184(2), 439-444. [17] Berg, W. A., Campassi, C., Langenberg, P., & Sexton, M. J. (2000). Breast Imaging Reporting and Data System: inter-and intraobserver variability in feature analysis and final assessment. American Journal of Roentgenology, 174(6), 1769-1777. [18] Obenauer, S., Hermann, K. P., & Grabbe, E. (2005). Applications and literature review of the BI-RADS classification. European radiology, 15(5), 1027-1036. [19] Elmore, J. G., Jackson, S. L., Abraham, L., Miglioretti, D. L., Carney, P. A., Geller, B. M., ... & Sickles, E. A. (2009). Variability in interpretive performance at screening mammography and radiologists’ characteristics associated with accuracy. Radiology, 253(3), 641-651. [20] White, E., Velentgas, P., Mandelson, M. T., Lehman, C. D., Elmore, J. G., Porter, P., ... & Taplin, S. H. (1998). Variation in mammographic breast density by time in menstrual cycle among women aged 40–49 years. Journal of National Cancer Institute, 90(12), 906-910. [21] Muttarak, M., Kongmebhol, P., & Sukhamwang, N. (2009). Breast calcifications: which are malignant?. Singapore medical journal, 50(9), 907-13. [22] Azavedo, E., & Bone, B. (1999). Imaging breasts with silicone implants. European radiology, 9(2), 349-355. [23] Song, QingZeng, et al. 'Using deep learning for classification of lung nodules on computed tomography images.' Journal of healthcare engineering 2017 (2017). [24] Doi, Kunio. 'Computer-aided diagnosis in medical imaging: historical review, current status and future potential.' Computerized medical imaging and graphics 31.4-5 (2007): 198-211. [25] Cao, Peng, et al. 'A multi-kernel based framework for heterogeneous feature selection and over-sampling for computer-aided detection of pulmonary nodules.' Pattern Recognition 64 (2017): 327-346. [26] Cheng, Jie-Zhi, et al. 'Computer-aided diagnosis with deep learning architecture: applications to breast lesions in US images and pulmonary nodules in CT scans.' Scientific reports 6 (2016): 24454. [27] Kuravatti, Rohit B., and B. Sasidhar. 'A Novel Method for Classification of Lung Nodules as Benign and Malignant using Artificial Neural Network.' International Journal of Engineering and Computer Science 3.08 (2014). [28] El-Baz, Ayman, et al. 'Computer-aided diagnosis systems for lung cancer: challenges and methodologies.' International journal of biomedical imaging 2013 (2013). [29] Hansell, David M., et al. 'Fleischner Society: glossary of terms for thoracic imaging.' Radiology 246.3 (2008): 697-722. [30] Fujimoto, Junya, and Ignacio I. Wistuba. 'Current concepts on the molecular pathology of non-small cell lung carcinoma.' Seminars in diagnostic pathology. Vol.31. No. 4. WB Saunders, 2014. [31] Gould, Michael K., et al. 'Accuracy of positron emission tomography for diagnosis of pulmonary nodules and mass lesions: a meta-analysis.' Jama 285.7 (2001): 914-924. [32] Gillies, Robert J., Paul E. Kinahan, and Hedvig Hricak. 'Radiomics: images are more than pictures, they are data.' Radiology 278.2 (2015): 563-577. [33] Zhao, Xinzhuo, et al. 'Agile convolutional neural network for pulmonary nodule classification using CT images.' International journal of computer assisted radiology and surgery (2018): 1-11. [34] Greenspan, Hayit, Bram van Ginneken, and Ronald M. Summers. 'Guest editorial deep learning in medical imaging: Overview and future promise of an exciting new technique.' IEEE Transactions on Medical Imaging 35.5 (2016): 1153-1159. [35] Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. 'Imagenet classification with deep convolutional neural networks.' Advances in neural information processing systems. 2012. [36] Simonyan, Karen, and Andrew Zisserman. 'Very deep convolutional networks for large-scale image recognition.' arXiv preprint arXiv:1409.1556 (2014). [37] Szegedy, Christian, et al. 'Going deeper with convolutions.' Cvpr, 2015. [38] He, Kaiming, et al. 'Deep residual learning for image recognition.' Proceedings of the IEEE conference on computer vision and pattern recognition. 2016. [39] Kotsiantis, Sotiris B., I. Zaharakis, and P. Pintelas. 'Supervised machine learning: A review of classification techniques.' Emerging artificial intelligence applications in computer engineering 160 (2007): 3-24. [40] Alpaydin, Ethem. Introduction to machine learning. MIT press, 2009. [41] I. Goodfellow, Y. Bengio, and A. Courville, Deep Learning. MIT Press, 2016, http://www.deeplearningbook.org. [42] S. Geman, E. Bienenstock, and R. Doursat, “Neural networks and the bias/variance dilemma”, Neural Computation, vol. 4, no. 1, pp. 1–58, 1992, ISSN: 1. [43] D. E. Rumelhart, G. E. Hinton, and R. J. Williams, “Learning representations by back-propagating errors”, Nature, vol. 323, no. 533, Oct. 1986. [44] V. Dumoulin and F. Visin, “A guide to convolution arithmetic for deep learning”, ArXiv e-prints, Mar. 2016. [45] Scherer, Dominik, Andreas Müller, and Sven Behnke. 'Evaluation of pooling operations in convolutional architectures for object recognition.' International conference on artificial neural networks. Springer, Berlin, Heidelberg, 2010. [46] Long, Jonathan, Evan Shelhamer, and Trevor Darrell. 'Fully convolutional networks for semantic segmentation.' Proceedings of the IEEE conference on computer vision and pattern recognition. 2015. [47] Cui, Yin, et al. 'Kernel pooling for convolutional neural networks.' Proceedings of the IEEE conference on computer vision and pattern recognition. 2017. [48]Bottou, Léon. 'Large-scale machine learning with stochastic gradient descent.' Proceedings of COMPSTAT'2010. Physica-Verlag HD, 2010. 177-186. [49] C. Muramatsu, T. Hara, T. Endo, H. Fujita, Breast mass classification on mammograms using radial local ternary patterns, Comput. Biol. Med. 72 (1) (2016) 43–53. [50] H. Li, X. Meng, T. Wang, Y. Tang, Y. Yin, Breast masses in mammography classification with local contour features, BioMed. Eng. Online 16 (1) (2017) 44– 54. [51] Z. Jiao, X. Gao, Y. Wang, J. Li, A deep feature based framework for breast masses classification, Neurocomputing 197 (C) (2016) 221–231. [52] N. Dhungel, G. Carneiro, A.P. Bradley, A deep learning approach for the analysis of masses in mammograms with minimal user intervention, Med. Image Anal. 37 (2017) 114–128. [53] G. Carneiro, J. Nascimento, A.P. Bradley, Automated analysis of unregistered multi-view mammograms with deep learning, IEEE Trans. Med. Image. 36 (11) (2017) 2355–2365. [54] Al-masni, Mohammed A., et al. 'Simultaneous detection and classification of breast masses in digital mammograms via a deep learning YOLO-based CAD system.' Computer methods and programs in biomedicine 157 (2018): 85-94. [55] Chokri, Ferkous, and Merouani Hayet Farida. 'Mammographic mass classification according to Bi-RADS lexicon.' IET Computer Vision 11.3 (2016): 189-198. [56] Heath, Michael, et al. 'The digital database for screening mammography.'Proceedings of the 5th international workshop on digital mammography. Medical Physics Publishing, 2000. [57] Szegedy, Christian, et al. 'Rethinking the inception architecture for computer vision.' Proceedings of the IEEE conference on computer vision and pattern recognition. 2016. [58] Pan, Sinno Jialin, and Qiang Yang. 'A survey on transfer learning.' IEEE Transactions on knowledge and data engineering 22.10 (2010): 1345-1359. | |
| dc.identifier.uri | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/21722 | - |
| dc.description.abstract | 根據全球癌症發病統計，乳癌位居女性癌症發生率第一位，也是在女性癌症中 的主要死因。最近醫療設備的進步以及乳房 X-ray 攝影技術的成熟，乳房中的微小 鈣化點、局部結構變形或腫塊(mass)等大多都能透過 Mammography 來偵測到，以 致越來越多沒有明顯症狀的早期乳癌能被提早發現。根據研究顯示，透過 Mammography 的篩檢來早期發現乳癌能有效降低 20%至 30%的乳癌死亡率。然 而，乳房組織的組成複雜，在不同放射科醫師間容易導致不同的判斷，難以用客觀 的方式評估。因此，為了讓乳房報告檢查有一致性，美國放射科醫學會發展 BI- RADS 書寫報告方式，即所謂的「乳房影像報告暨資料分析系統」，將 X-Ray 攝影 檢查結果分為七類。然而，目前根據 BI-RADS 對乳房病變的評估仍然大多基於定 性的描述以及主觀的判斷，還是具有相當的變異性，導致分級結果有較高的觀察者 間、內的變異。為了提高乳房 X-ray 攝影結果的準確性和一致性，許多研究團隊嘗 試使用機器學習方法來構建電腦輔助診斷 CAD 系統來進行定量的評估。目前 Mammography 相關的研究成果大多限於針對特定病例並且僅將病變區分為良性或 惡性，然而，一個適當的 CAD 系統必須根據臨床程序去設計，提供與放射科醫師 相同方式的輸出結果才能更佳 CAD 輔助醫療診斷的性能。
本研究提出使用貝氏定理的方式克服現有資料不足之問題，並將乳房 X-ray 攝 影中的腫瘤病變分類為 BI-RADS 3 4 5。在所有 BI-RADS 類別中，BI-RADS 3 4 5 的結果尤為重要。正確的判別 BI-RADS 3 4 5 不僅可以幫助早期篩檢和治療，還可 以避免不必要的切片檢查和手術。由於深度學習需要大量的訓練樣本，因此本研究 基於貝氏定理的方式，將已知的結果輸出透過前機率去推理每一類別的事後機率 來確定良惡性信息是否有助於 BI-RADS 的分類。在模型訓練過程中，image augmentation 以及 transfer learning 等常見的方法也被用於幫助避免 overfitting。本 研究針對 VGG16，ResNet50，DenseNet121 和 Inception-V3 等模型進行了測試，以 確定它們在良惡性和 BI-RADS 分類上的表現。對於良惡性腫瘤分類，Inception-V3 的表現優於其他網絡，總體準確度為 0.854，靈敏度為 0.843，特異性為 0.863。對 於 BI-RADS 分類，Inception-V3 的性能也優於其他網絡，總體準確率為 0.622。由於 Inception-V3 的表現在良惡以及 BI-RADS 的分類都優於其他網路，本研究以 Inception-V3 的結構去改良，透過貝氏定理將先驗機率使良惡性的訊息幫助正規化 訓練過程。這方法將BI-RADS分類accuracy 提升了10%，最終總體準確度為0.726， confusion matrix 顯示 BI-RADS 3 的敏感性為 0.701，BI-RADS 為 4:0.761，BI- RADS 為 5:0.717 。最後使用 CAM 顯現模型改良前後的熱度圖能夠看出加入先 驗機率去訓練能有效的改進模型提取重要的特徵像素。為輔助放射科醫師進行 BI-RADS 分類，本研究利用貝氏定理的方式將良惡性 訊息加入模型的訓練，克服現有的資料不足之問題。 由 confusion matrix 以及 CAM 皆顯示使用貝氏定理結合先驗知識可以幫助提高 BI-RADS 分類結果。 | zh_TW |
| dc.description.abstract | According to Global Cancer Statistics, breast cancer has been the most commonly diagnosed cancer and also the leading cause of cancer death among females. Recent improvements in medical technology and mammography show that early detections of microcalcifications, structural abnormalities and mass can be identified through mammograms. Studies indicate early detection can effectively reduce breast cancer mortality rate by 20 to 30%. However, it is difficult for radiologists to make consistent and objective evaluations. Consequently, in order to standardize image reporting and reduce confusion in breast imaging interpretations among radiologist, the American College of Radiology established a Breast Image Reporting and Data-analyzing System (BI-RADS), classifying lesions into 7 categories. However, current assessment of breast lesions according to BI-RADS remain qualitative and subjective, with substantial inter and intra-reader variability. In order to improve the accuracy and consistency of mammogram results, many studies have been conducted to build computer-aided diagnosis (CAD) systems using machine learning methods. However, developments and performance of latest classification systems are mostly limited to targeting specific cases and only differentiating lesions into benign or malignant, thus a proper suitable CAD system that outputs results according to BI-RADS with the same manner proceeded as radiologists is required to provide a second opinion in clinical settings.
In this study, novel approaches to classify mass lesions into BI-RADS 3 4 5 in mammograms are explored. Among all the BI-RADS categories, consistent results for BI-RADS 3 4 5 is particularly important. Proper reporting of BI-RADS 3 4 5 can not only help early detection and treatment, but also avoid unnecessary biopsies and surgeries. Since deep learning requires large number of training samples, a Bayesian framework has been investigated to see if incorporating malignancy information can help BI-RADS classification performance. Popular techniques such as data augmentation and transfer learning were also used to help avoid overfitting. State-of-the-art models such as VGG16, ResNet50, DenseNet121, and Inception-V3 were tested to determine each of their performance on both malignancy and BI-RADS classifications. For malignancy classification, Inception-V3 outperformed the rest of the networks with an overall accuracy of 0.854, sensitivity of 0.843 and specificity of 0.863. For BI-RADS classification, Inception-V3 also outperformed other networks with an overall accuracy of 0.622. The trained Inception-V3 network was later used as base model and fine-tuned with prior knowledge through a Bayesian framework to regularize the training process with malignancy estimates. This novel approach increased BI-RADS classification performance by 10% with a final overall accuracy of 0.726, the confusion matrix showed sensitivity of BI-RADS 3: 0.701, BI-RADS 4: 0.761 and BI-RADS 5: 0.717. Class activation maps also helped indicate an improvement in localization during prediction. With limited data, the results show that by using a Bayesian approach to incorporate prior- knowledge can help improve the performance of BI-RADS classification. | en |
| dc.description.provenance | Made available in DSpace on 2021-06-08T03:44:01Z (GMT). No. of bitstreams: 1 ntu-108-R06548059-1.pdf: 5763123 bytes, checksum: 73f3c4f7d7dc5283fcd9e87598c617e3 (MD5) Previous issue date: 2019 | en |
| dc.description.tableofcontents | Chapter 1 1
Introduction 1 1.1 Background 1 1.2 Objective 9 1.3 Thesis Overview 10 Chapter 2 11 Machine Learning 11 2.1 Overview of Machine Learning 11 2.1.1 Supervised Learning 12 2.1.2 Unsupervised learning 12 2.1.3 Overfitting 13 2.2 Neural Networks 14 2.2.1 Artificial Neural Network 14 2.2.2 Back Propagation 16 2.2.3 Cost Functions 17 2.3 Convolutional Neural Networks 17 2.3.1 Convolutional Layer 18 2.3.2 Pooling Layer 18 2.3.3 Fully Connected Layer 19 2.3.4 Global Average Pooling 19 2.4 Optimizations in Deep Learning 20 2.4.1 Gradient Descent 20 2.4.2 Batch Gradient Descent 21 2.4.3 Stochastic Gradient Descent 22 Chapter 3 23 Related Work 23 3.1 Malignancy Classification Models for Mammogram Mass 23 3.2 BI-RADS Classification Models for Mammogram Mass 25 Chapter 4 26 Materials & Methods 26 4.1 Materials 27 4.2 Methods 28 4.2.1 Mass Segmentation 29 4.2.2 Augmentation 30 4.2.3 Inception-V3 30 4.2.4 Transfer Learning 33 4.2.5 Malignancy Classification Model 34 4.2.6 BI-RADS Classification Model 35 4.2.7 BI-RADS Classification with Bayesian 35 4.2.8 Visualizing Deep Learning Models 40 Chapter 5 42 Experimental Results & Discussion 42 5.1 Evaluation Metric 42 5.2 Malignancy Classification Performance 43 5.3 BI-RADS 3 4 5 Classification Performance 44 5.4 BI-RADS 3 4 5 Classification with Bayesian 47 5.5 Discussion 50 Chapter 6 52 Conclusion 52 References 54 | |
| dc.language.iso | en | |
| dc.title | 基於貝氏定理以及深度學習針對 乳房攝影腫瘤進行 BI-RADS 分類 | zh_TW |
| dc.title | A Bayesian Approach to BI-RADS Classification of Mammogram Mass with Deep Learning | en |
| dc.type | Thesis | |
| dc.date.schoolyear | 107-2 | |
| dc.description.degree | 碩士 | |
| dc.contributor.oralexamcommittee | 張允中,王靖維,李佳燕 | |
| dc.subject.keyword | 乳房攝影良惡性分類,深度學習,卷積神經網路,資料增強, | zh_TW |
| dc.subject.keyword | BI-RADS,Bayesian,Deep Learning,Convolutional Neural Network,Data Augmentation,Class Activation Map, | en |
| dc.relation.page | 60 | |
| dc.identifier.doi | 10.6342/NTU201900753 | |
| dc.rights.note | 未授權 | |
| dc.date.accepted | 2019-05-10 | |
| dc.contributor.author-college | 工學院 | zh_TW |
| dc.contributor.author-dept | 醫學工程學研究所 | zh_TW |
| 顯示於系所單位： | 醫學工程學研究所 | |
文件中的檔案：
| 檔案 | 大小 | 格式 | |
|---|---|---|---|
| ntu-108-1.pdf 未授權公開取用 | 5.63 MB | Adobe PDF |
系統中的文件，除了特別指名其著作權條款之外，均受到著作權保護，並且保留所有的權利。