臺灣博碩士論文加值系統

乳癌是全球女性健康最大隱憂其中之一，在全世界女性死亡率排名為第二名，而在台灣女性死亡率為第一，發生高峰約在45-69歲之間。各期乳癌病患接受治療後存活率分別為下列，第零期為97.7％，第一期為95.7％，第二期為89.1％，第三期為72.3％，第四期為25.7％，若能越早發現則治療後存活率越高。中醫主要分為「望、聞、問、切」四診，而「舌診」屬於四診中的「望診」，中醫舌診亦隨著現代電腦資訊科技與知識經濟發展，在實務、研究、教學上有新的變革，強化舌診的判讀，將傳統中醫觀察的舌苔、舌色、舌質、舌下絡脈、朱點、瘀斑及齒痕等變化，經由圖像的自動判讀分析，提供檢測者體質虛實的現代診察儀器。所以我們希望利用統計方法來判斷中醫舌診資料是否能夠將資料進行分群，以便日後進行乳癌的預防與治療。在本文中我們先使用Gap statistic方法將資料進行分群數預測，依序利用Monte Carlo Sample=50、100、150、200，去探討其花費時間以及Gap的值。接著依照Gap statistic告訴我們的最佳分群數目，再去做k-means分群，此研究中我們發現Gap statistic可以快速且正確的找出最佳分群數目。也利用變異數分析去探討其分群的依據為何，最後再做多維度尺度分析(MDS)將多維資料投影到平面資料觀察其分群與乳癌其數的關係。

So far, as being ranked the second in worldwide female mortality and the first in Taiwan’s female mortality, breast cancer is among the serious problems of the female health in the world. It occurs most frequently during the age of forty five and sixty nine. The survival rates for all stages of breast cancer, after undergoing regular medical treatments, are 97.7%, 95.7%, 89.1%, 72.3%, and 25.7%, respectively for stage zero to stage four. In this thesis, we consider investigating a possible approach on classifying the stages of breast cancer by traditional Chinese medical tongue inspection. Apart from two papers by Cheng et al. (2014a,b) which demonstrates how to do clustering and predict the probability of occurrence of breast cancer , in this thesis we try to determine a proper number of clustering by Gap statistics before all clustering were conducted. In order to judge if Gap statistics is good in determining the number of staging by tongue inspection, we conduct data analysis for the sample obtained by the department of Traditional Chinese Medicine, Changhua Christian Hospital, Taiwan, to see if it is necessary and good before clustering such as k-means, multiple dimensional scaling (MDS), etc.

摘要 i
Abstract ii
致謝 iii
目錄 iv
圖目錄 v
表目錄 vi
第一章 緒論 1
第一節 研究動機及問題描述 1
第二節 研究對象 2
第二章 統計方法 4
第一節 分群的類型 4
第二節 本論文使用之分群的方法與介紹 5
A.間隙統計Gap statistic 5
B.k-平均算法(k-means) 7
C.多維度尺度分析(MDS) 9
第三章 主要結果與實例分析 11
第四章 結論 14
參考文獻 27

1.國民健康署乳癌防治文章(2013) ‧ 取自http://www.hpa.gov.tw/BHPNet/Web/healthtopic/TopicArticle.aspx?No=201312230001&parentid=200712250033
2.林震岩(2012)‧多變量分析:spss的操作與應用‧台北:智勝文化。
3.陳景祥(2012)‧R軟體:應用統計方法‧台北:東華書局。
4.羅綸謙, 蔣依吾(2010)‧臨床望舌彩色圖解。
5.廖偉志(2014碩士論文) ‧使用二階段虛擬最大概似估計法預測乳癌分期之探究‧國立彰化師範大學統計資訊研究所。
6.林逸捷(2014碩士論文) ‧根據舌象資料利用多維度分群作乳癌分期‧國立彰化師範大學數學所。
7.Anderberg, M.R. (1973). Cluster Analysis for Applications. Academic Press.
8.Borg, I., Groenen, P. (1997). Modern Multidimensional Scaling, Theory and Applications. Springer-Verlag, New York.
9.Cox, T.F., Cox, M.A.A., (1994). Multidimensional Scaling. Chapmanand Hall, London.
10.Gilks, W., Richardson, S., & Spiegelhalter, D. (1995). Markov Chain Monte Carlo in practice. London, UK: Chapman and Hall.
11.Huang, Z.( 1997a). Clustering large data sets with mixed numeric and categorical values. Proceedings of the First Pacific Asia Knowledge Discovery and Data Mining Conference, Singapore: World Scientific, pp. 21–34.
12.Huang, Z.( 1997b). A fast clustering algorithm to cluster very large categorical data sets in data mining. Proceedings of the SIGMOD Workshop on Research Issues on Data Mining and Knowledge Discovery, Dept. of Computer Science, The University of British Columbia, Canada, pp. 1–8.
13.Huang, Z. (1998). Extensions to the k-means algorithm for clustering large data sets with categorical values. Data Mining and Knowledge Discovery 2, pp.283-304.
14.Kruskal, J.B.,(1964). Nonmetric multidimensional scaling: A numerical method.Psychometrika 29,pp. 115–129.
15.Kaufman, L. and Rousseeuw, P.J. 1990. Finding Groups in Data—An Introduction to Cluster Analysis. Wiley.
16.Li, Z., Yuan, J., Yang, H., & Zhang, K (2008). K-mean Algorithm with a Distance Based on the Characteristic of Differences. In Wireless Communications, Networking and Mobile Computing, 2008. WiCOM'08. 4th International Conference on (pp. 1-4). IEEE.
17.Lo, L. C., Chen, Y. F., Chen, W. J., Cheng, T. L., Chiang, J. Y. (2012). The study on the agreement between automatic tongue diagnosis system and traditional chinese medicine practitioners. Evidence-Based Complementary and Alternative Medicine,2012.
18.Lo, L. C., Cheng, T. L., Chiang, J. Y., Damdinsuren, N. (2013). Breast Cancer Index:A Perspective on Tongue Diagnosis in Traditional Chinese Medicine. Journal of Traditional and Complementary Medicine, Vo1. 3, No. 3, pp. 194-203.
19.Moldovanu, S., Moraru, L. (2010). Mass detection and classification in breast ultrasound image using K-means clustering algorithm. In Electrical and Electronics Engineering (ISEEE), 2010 3rd International Symposium on (pp. 197-200). IEEE.
20.Roman Rouzier, Charles M. Perou, W. Fraser Symmans, et al. (2005). Breast cancer molecular subtypes respond differently to preoperative chemotherapy.Clinical Cancer Research, 2005;11,pp. 5678-5685.
21.Shabbar Naqvi, Jonathan M. Garibaldi.(2011). complexities involved in the analysis of Fourier Transform Infrared Spectroscopy of breast cancer data with clustering algorithms.In Computer Science and Electronic Engineering Conference (CEEC), 2011 3rd (pp. 80-85). IEEE.
22.Selim, S.Z. and Ismail, M.A. (1984). k-Means-type algorithms: A generalized convergence theorem and characterization of local optimality. IEEE Transactions on Pattern Analysis and Machine Intelligence 6,pp.81–87.
23.Tibshirani, R. (2001).Estimating the Number of Clusters Via the Gap Statistic.
24.Zhong, W., Altun, G., Harrison, R., Tai, P. C., Pan, Y. (2005). Improved K-Means Clustering Algorithm for Exploring Local Protein Sequence Motifs Representing Common Structural Property. NanoBioscience, IEEE Transactions on, VOL. 4, NO. 3,pp.255-265.