臺灣博碩士論文加值系統

本研究在於發展一子宮頸抹片影像為基礎的診斷系統(Image based Liquid-Based Smear Diagnosis System, ILSD system)。
細胞質與細胞核的大小、形狀、顏色深淺與內部紋路(texture)等，跟子宮頸細胞正常與否有著密不可分的關係。如何從子宮頸細胞抹片影像中，精確地切割出細胞質與細胞核的區域，是子宮頸細胞抹片電腦自動化篩檢系統中，最重要的關鍵技術之ㄧ，學術上已有許多人提出相關研究及方法。因此，本研究將著重在細胞辨識之工作，發展一子宮頸抹片影像為基礎診斷系統(Image based Liquid-Based Smear Diagnosis System, ILSD system)。子宮頸抹片影像細胞可能為正常細胞的淺層細胞、中層細胞、側基底細胞，以及異常細胞的低度鱗狀上皮細胞病變、高度鱗狀上皮細胞病變、鱗狀細胞癌等，而本研究所提之ILSD system則是利用子宮頸抹片病理特徵，如細胞的形狀、細胞核與細胞質之面積比例(核質比)、核膜之規則性、染色質之含量及分布等特徵，判斷子宮頸抹片影像細胞是何種細胞，以供醫師作為診療之參考。為使上述各項工作能獲得最佳效能，本研究同時也提出一基因演算法，來決定上述各項工作所使用到參數之最適當值。
本研究雖能對含單一子宮頸抹片細胞影像，進行精確辨識，但部份子宮頸癌症，無法純粹從單一細胞之特徵來判定，需結合細胞群聚狀況才能判定。因此未來研究應將細胞群聚特徵也列入考慮。

In this study, we develop an image based Liquid-based smear diagnosis system (ILSD system).The size, shape, color depth and internal texture of the cytoplasm and nucleus have a close relationship with the cervical cells which are normal or not. Segment the cytoplasm and nucleus region accurately from the cervical smear image is the most important technology in the automatic screening cervical smear system. Many scholars have proposed relevant research and methods. Therefore, this study proposed an image based Liquid-based smear diagnosis system (ILSD system) which focuses on cell identification. Cervical smear cells may be superficial squamous cell (S Cell), intermediate squamous cell (I Cell), parabasal squamous cell (P Cell ), Low-grade squamous intraepithelial lesion cell (LSIL Cell), or high-grade squamous intraepithelial lesion cell (HSIL Cell), etc.
The ILSD system used pathological features of the cervical smear, such as shape of cell, the ratio of nucleus’ area to the cytoplasm’s area, the regularity of the nuclear membrane, the content of the chromatin and distribution of the chromatin to provide information of treatment to doctor as a reference. Furthermore, a genetic algorithm is proposed to achieve the best performances to determine the most appropriate parameters in the each task. Although this study can accurately identify single cervical smear images, some of the cervical cancer can not be determined purely from the characteristics of a single cell. It needs to combine the cell cluster to determine the situation

摘要 i
Abstract ii
Table of Contents iii
List of Tables iv
List of Figures v
Chapter 1 Introduction 1
Chapter 2 Related Works 8
2.1. Cell Classification 8
2.2. Semi-Automatic Segmentation and Classification of Pap Smear Cells 12
Chapter 3 Cervical Smear Cell Image Recognition 16
3.1. Cervical Cell Feature Extraction 16
3.2. Classification of Cervical Cells 20
3.3.Genetic-Based Parameter Detector (GBPD) 23
Chapter 4 Experimental Results and Discussions 26
4.1. The Performance Test of Image Based Liquid-Based Smear Diagnosis System 26
4.2. The Comparison of the Classification Accuracy of Cervical Cells 29
Chapter 5 Conclusion and Future Works 31
Reference 33

[1]黃博琪(2003)，”子宮頸細胞自動化電腦輔助影像分析系統之設計研發”，中興大學生物醫學研究所碩士論文
[2]陳淑嬌(2003)，“梯度向量流主動輪廓模型於子宮頸抹片細胞邊界偵測之應用”，中興大學電機工程研究所碩士論文
[3]行政院衛生署(2003)，臺灣地區歷年癌症死亡數統計http://www.doh.gov.tw/statistic/死因統計/st4_91_1.htm
[4]嚴孟祿與謝長堯(2001)，子宮頸癌在台灣，行政院衛生署80年度委託研究計畫成果報告
[5]長庚紀念醫院曾志仁醫師，”子宮頸癌與子宮頸癌前期的防治” 中華民國衛生署子宮頸癌防治教材
http://www.csh.org.tw/dr.tcj/Tseng2013/html/education-prevention.html
[6]鐘坤井醫師，”細胞病理學”
http://www.cgmh.com.tw/intr/intr5/ c6700/teaching /professional/婦科細胞病理學.htm.
[7]A.M. Andrew. "Another efficient algorithm for convex hulls in two dimensions." Information Processing Letters 9,5 (1979): 216-219.
[8]N. A. Mat-Isa, M. Y. Mashor, and N. H. Othman. "Seeded region growing features extraction algorithm; its potential use in improving screening for cervical cancer." International Journal of The Computer, the Internet and Management 13,1 (2005): 61-70.
[9]A. P. Bradley, G. Carneiro, Z. Lu, and N. Dhungel. "Segmentation of overlapping cervical cells from multi-layer cytology preparation volumes." International Symposium on BIOMEDICAL IMAGING: From Nano to Macro, April 16-19 (2015).
[10]T. Chankong , N. Theera-Umpon , and S. Auephanwiriyakul. "Automatic cervical cell segmentation and classification in Pap smears." Computer methods and programs in biomedicine 113,2 (2014): 539-556.
[11]K. Li , Z. Lu, W. Liu , and J. Yin. "Cytoplasm and nucleus segmentation in cervical smear images using Radiating GVF Snake." Pattern Recognition 45,4 (2012): 1255-1264.
[12]A. H. Mbaga, and P. ZhiJun. "Pap Smear Images Segmentation for Automatic Detection of Cervical Cancer." cell 9,1 (2013):940-943.
[13]S. F. Yang-Mao, Y. K. Chan, and Y. P. Chu. "Edge enhancement nucleus and cytoplast contour detector of cervical smear images." IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics) 38,2 (2008): 353-366.
[14]R. Nayar, and D. C. Wilbur. "The Bethesda system for reporting cervical cytology: definitions, criteria, and explanatory notes." Springer (2015).
[15]Y. F. Chen, P. C. Huang, K. C. Lin, H. H. Lin, L. E. Wang, C. C. Cheng, T. P. Chen, Y. K. Chan, and J. Y. Chiang. "Semi-automatic segmentation and classification of pap smear cells." IEEE Journal of Biomedical and Health Informatics 18,1 (2014): 94-108.
[16]R. M. Haralick, and K. Shanmugam. "Textural features for image classification." IEEE Transactions on systems, man, and cybernetics 3,6 (1973): 610-621.
[17]H. Tamura, S. Mori, and T. Yamawaki. "Textural features corresponding to visual perception." IEEE Transactions on Systems, Man, and Cybernetics 8,6 (1978): 460-473.
[18]S. Yonekawa, N. Sakai, and O. Kitani. "Identification of idealized leaf types using simple dimensionless shape factors by image analysis." Transactions of the ASAE-American Society of Agricultural Engineers 39,4 (1996): 1525-1534.
[19]H. C. Lin, C. Y. Chiu, and S. N. Yang. "Finding textures by textual descriptions, visual examples, and relevance feedbacks." Pattern Recognition Letters 24,14 (2003): 2255-2267.
[20]R. C. Gonzalez, and R. E. Woods. "Digital Image Processing." Addison-Wesley Longman Publishing Co., Inc. (2002).
[21]B. Pathak, and D. Barooah. "Texture analysis based on the gray-level co-occurrence matrix considering possible orientations." International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering 2,9 (2013): 4206-4212.
[22]J. MacQueen. "Some methods for classification and analysis of multivariate observations." Proceedings of the fifth Berkeley symposium on mathematical statistics and probability. 1,14 (1967).