進階搜尋


下載電子全文  
系統識別號 U0026-1508201920034800
論文名稱(中文) 基於彩色和紋理圖像的深度卷積神經網絡口腔癌病兆分割
論文名稱(英文) Oral Cancer Lesion Segmentation Using Deep Convolution Neural Network Based on Color and Texture Images
校院名稱 成功大學
系所名稱(中) 電腦與通信工程研究所
系所名稱(英) Institute of Computer & Communication
學年度 107
學期 2
出版年 108
研究生(中文) 黃士洋
研究生(英文) Shih-Yang Huang
學號 Q36061096
學位類別 碩士
語文別 英文
論文頁數 62頁
口試委員 指導教授-詹寶珠
口試委員-陳智楊
口試委員-曾盛豪
口試委員-黃則達
口試委員-張建緯
中文關鍵字 口腔癌  小波轉換  賈伯濾波器  深度卷積神經網路 
英文關鍵字 oral cancer  wavelet transform  Gabor filter  convolutional network 
學科別分類
中文摘要 全世界據估計每年增加口腔癌657,000新案例,死亡人數超過33萬人次。如果在早期診斷出口腔癌,則整體5年生存率高達84%。然而如果癌症已擴散到身體的遠處部位,那麼5年生存率僅剩下39%。因此,在早期階段檢測口腔癌至關重要。與侵入性方法相比,視覺無創檢查是進行口腔癌初步診斷的更有效和可行的方法。因此,該研究應用幾種深度卷積神經網路(CNN)來分割口腔中的癌症和癌前病變區域。所提出的CNN都基於完全卷積網絡(FCN)結構,其中包含四個不同的子模塊,即特徵金字塔網路(FPN),擠壓並激發網路(SE-net),U-net的堆疊概念和聚合節點(Aggregation nodes)。對於每個CNN,輸入圖像包括RGB彩色圖像以及通過小波變換獲得的四組特徵圖,或者使用賈伯濾波器所獲得的四個特徵圖。此研究使用交集聯合(IOU),靈敏度,特異性和準確度度量來評估所提出的深度學習框架的可行性。初步實驗結果表明,IOU,靈敏度和特異性分別達到0.522, 0.78, 0.75。
英文摘要 An estimated 657,000 new cases of oral cavity cancer every year in the world, and more than 330,000 deaths. If oral cancer is diagnosed at an early stage, the overall 5-year survival rate is as much as 84%. However, if the cancer has spread to distant parts of the body, the 5-year survival rate is just 39%. Thus, it is essential to detect oral cancer at the earlier stage possible. Visual non-invasive examination is a more efficient and feasible approach for performing a preliminary diagnosis of oral cancer than invasive approaches. Accordingly, this study applies several deep convolutional neural networks (CNNs) to segment cancer and precancer lesion regions in the oral cavity. The proposed CNNs are all based on a fully convolutional network (FCN) structure, but incorporate four different sub-modules, namely FPN, SE-net, concatenate concept of U-net, and aggregation node. For each CNN, the input images include the RGB color image, four sets of feature maps obtained by wavelet transform, or four feature maps obtained using a Gabor filter. The feasibility of the proposed framework is evaluated using the intersection over union (IOU), sensitivity, specificity metrics. The preliminary results show that the IOU, sensitivity, and specificity achieve 0.522, 0.78, 0.75 respectively.
論文目次 摘 要 II
Oral Cancer Lesion Segmentation Using Deep Convolution Neural Network Based on Color and Texture Images IV
Abstract IV
List of Tables VIII
List of Figures IX
Chapter 1 Introduction 1
Chapter 2 Related Works 5
Chapter 3 Materials and Methods 10
3.1 Oral Cancer Detection System Using Color and Auto-fluorescence Images 10
3.1.1 Oral Cancer Detection Algorithm 10
3.2 Texture Feature Map Extraction 13
3.2.1 Texture Feature 13
3.2.2 2D Discrete Wavelet Transformation 16
3.2.3 2D Gabor Filter 20
3.3 Oral Cancer Detection Using Deep Learning Methods 23
3.3.1 Deep Learning Using GPUs 23
3.3.2 Semantic Segmentation 23
3.3.3 Fully Convolutional Network 24
3.3.4 The Concatenate Concept of U-Net 29
3.3.5 Feature Pyramid Network (FPN) 32
3.3.6 Squeeze-and-Excitation Network 35
3.3.7 The Model with Aggregation Nodes 38
Chapter 4 Experimental Results and Discussions 40
4.1 Oral Cavity Color Image dataset 40
4.2 Evaluation Criterion 41
4.3 Training Configuration 43
4.3.1 Accuracy of Networks 43
4.4 Results of Marking ROI 46
4.4.1 Normal case results 46
4.4.2 Precancer case results 49
4.4.3 Cancer case results 52
Chapter 5 Conclusion 55
Reference 57
參考文獻 1. The Oral Cancer Foundation Oral Cancer Facts, U.S.A., Available at:
https://oralcancerfoundation.org/facts/
2. Health Promotion Administration, Ministry of Health and Welfare, Taiwan. Cancer Prevention and Control. Available at:
https://www.hpa.gov.tw/Pages/Detail.aspx?nodeid=613&pid=1118
3. P. M. Williams, C. F. Poh, A. J. Hovan, S. Ng, and M. P. Rosin., "Evaluation of a suspicious oral mucosal lesion," J. Can. Dent. Assoc., vol. 74, no. 3, pp. 275-280, 2008.
4. B. W. Neville and T. A. Day, "Oral cancer and precancerous lesions," C. A. Cancer J. Clin., vol. 52, no. 4, pp. 195-215, Jul. - Aug. 2002.
5. T. W. Remmerbach, N. Pomjanski, and U. Bauer, H. Neumann. "Liquid-based versus conventional cytology of oral brush biopsies: a split-sample pilot study, " Investig. Clin. Oral., vol. 21, no. 8, pp. 2493-2498, Jan. 2017.
6. Wang Xiaoqian, Elżbieta Kaczor-Urbanowicz Karolina, and T.W. Wong David, "Salivary Biomarkers in Cancer Detection," Med. Oncol., vol. 34, no. 1, pp. 7, Jan. 2017.
7. Y. W. Chen et al., "Use of methylene blue as a diagnostic aid in early detection of oral cancer and precancerous lesions," Br. J. Oral Maxillofac. Surg., vol. 45, no. 7, pp. 590-591, Nov. 2007.
8. S. Dongsuk, V. Nadarajah, G. Ann, and R.-K. Rebecca, "Advances in fluorescence imaging techniques to detect oral cancer and its precursors," Future Oncol., vol. 6, no. 7, pp. 1143-1154, Jul. 2010.
9. S. Arivazhagana and L. Ganesan, "Texture classification using wavelet transform," Pattern Recognition Letters., vol. 24, no.9-10, pp. 1513-1521, Jun. 2003.
10. S. Livens, P. Scheunders, G. van de Wouwer, and D. Van Dyck, "Wavelets for texture analysis, an overview," 6th International Conference on Image Processing and its Applications., pp. 581-585, 1997.
11. Omar S. AI-Kadi, "A Gabor Filter Texture Analysis Approach for Histopathological Brain Tumor Subtype Discrimination," ISESCO., vol. 12, no. 22, Apr. 2003.
12. J.G. Daugman, "Complete discrete 2-D Gabor transforms by neural networks for image analysis and compression," IEEE Transactions on Acoustics, Speech, and Signal Processing., vol. 36, no. 7, pp. 1513-1521, Jul. 1998.
13. P. Kruizinga, N. Petkov, " Nonlinear operator for oriented texture," IEEE Transactions on Image Processing., vol. 8, no. 10, pp. 1395-1407, Oct. 1999.
14. T. Mihran and A. K. Janil, "TEXTURE ANALYSIS," Handbook of Pattern Recognition and Computer Vision., pp. 235-276, 1993
15. T. Chang and C.-C.J. Kuo, "Texture analysis and classification with tree-structured wavelet transform," IEEE Transactions on Image Processing., vol. 2, no. 4, pp. 429-441, Oct. 1993.
16. S. Arivazhagan, L. Ganesan, and V. Angayarkanni, "Color texture classification using wavelet transform," ICCIMA 2005., Nov. 2005.
17. G. Castellano, L. Bonilha, L.M. Li, and F.Cendes, "Texture analysis of medical images," Clinical Radiology., vol. 59, no. 124, pp. 1061-1069, Dec. 2004.
18. S. E. Grigorescu, N. Petkov, and P. Kruizinga, "Comparison of texture features based on Gabor filters," IEEE Transactions on Image Processing., vol. 11, no. 10, pp. 1160-1167, Dec. 2002.
19. H. Yoshihiko, U. Shunju, W. Masanori, Y. Tetsuya, M. Yoshihiro, and T. Shingo, "A gabor filter-based method for recognizing handwritten numerals," Pattern Recognition., vol. 31, no. 4, pp. 395-400, Apr. 1998.
20. J. V. B. Soares, J. J. G. Leandro, R. M. Cesar, H. F. Jelinek, and M. J. Cree, "Retinal vessel segmentation using the 2-D Gabor wavelet and supervised classification," IEEE Transactions on Medical Imaging., vol. 25, no. 9, pp. 1214-1222, Sept. 2006.
21. A. Krizhevsky, I. Sutskever, and G. E. Hinton, "ImageNet classification with deep convolutional neural networks," NIPS 2012., vol. 1, pp. 1097-1105, Dec. 2012.
22. K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” ICLR 2015, May 2015.
23. R. Girshick, J. Donahue, T. Darrell, and J. Malik, "Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation," CVPR 2014., Jun. 2014.
24. R. Girshick, "Fast R-CNN," ICCV 2015., pp. 1440-1448, Dec. 2015.
25. S. Ren, K. He, G. Girshick, and J. Sun, "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks," IEEE Transactions on Pattern Analysis and Machine Intelligence., vol. 39, no. 6, pp. 1137-1149, Jun. 2016.
26. J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, "You Only Look Once: Unified, Real-Time Object Detection," CVPR 2016., Jun. 2016.
27. J. Redmon and A. Farhadi, "YOLO9000: Better, Faster, Stronger," CVPR 2017., Nov. 2017.
28. I. J. Goodfellow et al, "Generative Adversarial Nets," NIPS 2014., vol. 2, pp. 2672-2680, Dec. 2014
29. P. Luc, C. Couprie, S. Chintala, and J. Verbeek, "Semantic Segmentation using Adversarial Networks," NIPS 2016., Dec. 2016
30. J.-Y. Zhu, T. Park, P. Isola, and A. A. Efros, “Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks,” The IEEE ICCV 2017., pp. 2223-2232, 2017
31. T. Karras, T. Aila, S. Laine, and J Lehtinen, “Progressive Growing of GANs for Improved Quality, Stability, and Variation,” ICLR 2018, Feb. 2018
32. Y. Li, S. Liu, J. Yang, and M. -H. Yang, “Generative Face Completion,” The IEEE CVPR 2017., pp. 3911-3919, 2017
33. M. Arjovsky, S. Chintala, and L. Bottou, “Wasserstein GAN,” ICML 2017., vol. 70, pp. 214-223, 2017
34. Xudong Mao et al, “Least Squares Generative Adversarial Networks,” The IEEE ICCV 2017., pp. 2794-2802, 2017
35. J. Long, E. Shelhamer, and Trevor Darrell, “Fully Convolutional Networks for Semantic Segmentation,” The IEEE CVPR 2015., pp. 3431-3440
36. O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional Networks for Biomedical Image Segmentation,” MICCAI 2015., pp. 234-241
37. V. Badrinarayanan, A. Kendall, and R. Cipolla, “SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation,” IEEE Transactions on Pattern Analysis and Machine Intelligence 2017., vol. 39, no. 12, pp. 2481 – 2495, Dec. 2017
38. K. He, J. Sun, “Convolutional Neural Networks at Constrained Time Cost,” The IEEE CVPR 2015., pp. 5353-5360, 2015
39. K. He, X. Zhang, S. Ren, and J. Sun; “Deep Residual Learning for Image Recognition,” The IEEE CVPR 2016., pp. 770-778, 2016
40. L.-C. Chen, G. Papandreou, I. Kokkinos, K. Murphy, and A. L. Yuille, “Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs,” ICLR 2015., vol. abs/1412.7062, 2015
41. L.-C. Chen, G, Papandreou, I. Kokkinos, K. Murphy, and A. L. Yuille, “DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs,” IEEE Transactions on Pattern Analysis and Machine Intelligence., vol. 40, no. 4, April 2018
42. L.-C. Chen, G. Papandreou, F. Schroff, and H. Adam, “Rethinking Atrous Convolution for Semantic Image Segmentation.” ArXiv 2017., abs/1706.05587, 2017
43. L.-C. Chen, Y. Zhu, G. Papandreou, F. Schroff, and H. Adam, “Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation,” ECCV 2018., pp. 801-818, 2018
44. F. Yu and V. Koltun, “Multi-Scale Context Aggregation by Dilated Convolutions,” ICLR 2016., 2016
45. H. Zhao, J. Shi, X. Qi, X. Wang, and J. Jia, “Pyramid Scene Parsing Network,” CVPR 2017., pp. 2881-2890, 2017
46. J. Hu, L. Shen, and G. Sun, “Squeeze-and-Excitation Networks,” CVPR 2018., pp. 7132-7141, 2018
47. C. Szegedy et al., “Going Deeper With Convolutions,” CVPR 2015., pp. 1-9, 2015
48. S. loffe and C. Szegedy, “Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift,” arXiv: 1502.03167 [cs], Mar. 2015.
49. C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, “Rethinking the Inception Architecture for Computer Vision,” CVPR 2016., pp. 2818-2826, 2016
50. C. Szegedy, S. Ioffe, V. Vanhoucke, and A. A. Alemi, “Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning,” AAAI., Feb. 2017
51. T.-Y. Lin, P. Dollar, R. Girshick, K. He, B. Hariharan, and S. Belongie, “Feature Pyramid Networks for Object Detection,” CVPR 2017., pp. 2117-2125, 2017
52. T.-Y. Huang, “Application of Autofluorescence Imaging On Clinical Oral Cancer Screening,” Available at:
http://etds.lib.ncku.edu.tw/etdservice/view_metadata?etdun=U0026-1508201423052400
53. F. Yu, D. Wang, and T. Darrell, “Deep Layer Aggregation,” CVPR 2018., pp. 2117-2125, Jun.2018
54. A. Chodorowski, U. Mattsson, T. Gustavsson, “Oral lesion classification using true-color images”, Proceedings of SPIE, Medical Imaging: Image Processing, K. M. Hansson, Editor, Vol. 3661, pp. 1127-1138, 1999.
論文全文使用權限
  • 同意授權校內瀏覽/列印電子全文服務,於2020-01-01起公開。
  • 同意授權校外瀏覽/列印電子全文服務,於2020-01-01起公開。


  • 如您有疑問,請聯絡圖書館
    聯絡電話:(06)2757575#65773
    聯絡E-mail:etds@email.ncku.edu.tw