跳到主要內容

臺灣博碩士論文加值系統

(18.97.14.86) 您好!臺灣時間:2025/02/07 22:51
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:胡芳瑋
研究生(外文):Fang-Wei Hu
論文名稱:一個使用小波及獨立成分分析診斷乳房X光影像微鈣化的電腦輔助診斷系統
論文名稱(外文):A computer aided diagnosis system for microcalcifications in digital mammograms based on wavelet and Independent Component Analysis
指導教授:余松年余松年引用關係
指導教授(外文):Sung-Nien Yu
學位類別:碩士
校院名稱:國立中正大學
系所名稱:電機工程所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2006
畢業學年度:95
語文別:中文
論文頁數:75
中文關鍵詞:小波轉換乳房微鈣化組織獨立成分分析
外文關鍵詞:microcalcificationwaveletindependent component analysis
相關次數:
  • 被引用被引用:3
  • 點閱點閱:406
  • 評分評分:
  • 下載下載:77
  • 收藏至我的研究室書目清單書目收藏:1
本論文主要提出一個應用於乳房X光影像中微鈣化組織的電腦輔助診斷系統,本方法主要分為兩階段的影像處理,包含了微鈣化組織的偵測與微鈣化組織的診斷。在第一階段我們根據乳房組織的病理特徵,將影像使用線性結構的方法找出乳管,再利用臨界值的選取留下可疑的候選區塊,將這些候選區塊作小波能量特徵擷取送入倒傳遞神經網路分類器分類,得到正常組織與微鈣化組織。第二階段,將第一階段測得的微鈣化組織,作特徵擷取,使用的方法為獨立成分分析法,根據取樣分析得到最佳的結果,將所有待分類的區塊投影至該獨立成分空間,利用支持向量機將微鈣化組織區分為良性與惡性。本論文採用MIAS資料庫。經由實驗結果顯示,在第一階段,本論文所提出的方法在432個誤報數中正確的去除了417個,降低的誤報率達96.53%,靈敏度為95.1%,平均每張影像的誤報數為0.75。第二階段,使用獨立成分分析在取樣部份正確率可高達100%,而實際應用至影像上,靈敏度為76%,特異度為100%,而整體正確率為85.37%,在乳房X光影像電腦輔助診斷系統中算是不錯的效能。
A two-stage computer aided diagnosis system for microcalcifications in digital mammograms is proposed in this thesis. The two stages include image processing for microcalcifications detection and diagnosis respectively. In the first stage, we classify normal from abnormal breast tissues. The linear structure method is used to find the ducts in the mammogram according to the pathology of mammograms. A threshold method is then employed to find suspicious blocks for further classification. Wavelet transform is used for feature extraction and the back propagation neural network is used for the classification of the suspicious blocks. In the second stage, the microcalcifications detected in the first stage are classified into benign or malignant. The features are extracted by independent component analysis method. We use the ROI (region of interest) samples to generate the ICA base. The abnormal image samples are projected onto the independent component bases, and then the support vector machine is used for classification. The MIAS database is used for experiments. In the first stage, our method can readily reduce the false positive from 432 to 15, resulting in a reducing rate of 95.53%. Beside, the sensitivity is 95.1%, with 0.75 false positives per image. In the second stage, the recognition accuracy by independent component analysis can reach 100% for the sample images. Further testing the diagnosis system with whole images, the system demonstrates a sensitivity of 76%, a specificity of 100%, and a high recognition accuracy of 85.37%, The results are quite impressive in a computer aided diagnosis system for digital mammograms.
摘要 1
Abstract 2
目錄 3
圖目錄 6
表目錄 8
第一章 導論 9
1.1 研究背景 9
1.1.1 乳癌病理學 9
1.1.2 乳癌的檢查與診斷 9
1.1.3 乳房微鈣化組織 10
1.1.4 乳房X光影像 11
1.2 研究動機與目的 13
1.2.1 研究動機 13
1.2.2 研究目的 13
1.3 研究方法 14
1.4 論文架構 15
第二章 相關文獻回顧 16
2.1 乳房影像增強 16
2.2 線性組織偵測 16
2.3 微鈣化組織的偵測 17
2.4 微鈣化組織的診斷 18
第三章 本研究相關方法 20
3.1 直方圖等化 20
3.2 線性結構 21
3.3 小波轉換 22
3.3.1 正交小波轉換 23
3.3.2 尺度函數 23
3.3.3 二維小波轉換 23
3.3.4 小波統計特徵 25
3.4 主成分分析 26
3.5 獨立成分分析 27
3.6 倒傳遞類神經網路 29
3.7 支持向量機 33
3.8 線性區別分類器 37
3.9 實驗數據正規化 37
3.10 實驗評估標準 38
第四章 系統設計與數據分析 39
4.1 系統架構 39
4.2 前處理 40
4.2.1 直方圖等化 40
4.2.2 線性結構 40
4.3 偵測階段 45
4.3.1 特徵擷取-小波分解 45
4.3.2 分類器-倒傳遞神經網路 47
4.3.3 結論 48
4.4 診斷階段 48
4.4.1 特徵擷取-小波分解 48
4.4.2 獨立成分分析 50
4.4.3 前處理-主成分分析 52
4.4.4 結合獨立成分特徵與小波特徵 54
4.4.5 分類器-支持向量機 57
4.4.6 分類器-線性區別分類器 58
4.4.7 結論 58
4.5 結語 59
第五章 系統測試與討論 60
5.1 偵測結果 60
5.2 診斷結果 66
第六章 結論與未來發展 71
6.1 結論 71
6.2 未來發展 71
參考文獻 73
[1]http://www.fma.org.tw/medicial_data/taiwan01.htm#07台灣醫學會,行政院衛生署八十二年度委託研究計畫研究報告
[2]http://www.libertytimes.com.tw/2005/new/oct/21/life/medicine-1.htm國泰醫院乳癌中心主任杜世興
[3]J Suckling et al. “The mammographic images analysis society digital mammogram database”, Exerpta Medica. International Congress Series, vol. 1069, pp. 375378, 1994
http://www.wiau.man.ac.uk/services/MIAS/MIASweb.html

[4]Fatima L. S. Nunes, Homero Schiabel et al. “ethod to contrast enhancement of digital dense breast images aimed to detect clustered microcalcifications” IEEE International Conference on Image Processing, v 1, 2001, p 305-308
[5]A.Raji et al. “A gray-level transformation-based method for image enhancement” Pattern Recognition Letters, v 19, n 13, Nov, 1998, p 1207-1212
[6]Reyer Zwiggelaar et al. “Linear structures in mammographic images: detection and classification” IEEE Transactions on Medical Imaging, v 23, n 9, September, 2004, p 1077-1086
[7]Song-yang Yu, Ling Guan “A CAD system for the automatic detection of clustered microcalcifications in digitized mammogram films” IEEE Transactions on Medical Imaging, v 19, n 2, Feb, 2000, p 115-126
[8]T. O. Gulsrud, J. H. Husoy, “Optimal Filter-Based Detection of Microcalcifications,” IEEE Trans. On Biomedical Engineering, vol. 48,
no. 11, Nov. 2001.
[9]El-Naqa et al. “A support vector machine approach for detection of microcalcifications” IEEE Transactions on Medical Imaging, v 21, n 12, December, 2002, p 1552-1563
[10]Ioanna Christoyianni et al, “Breast tissue classification in mammograms using ICA mixture models” ICANN 2001,LNCS2130,pp. 554-560,2001
[11]P. Comon, “ Independent component analysis. A new concept?” Signal Processing, v 36, n 3, Apr, 1994, p 287-314
[12]http://www.cis.hut.fi/

[13]A. Hyvarinen,J. Karhunen, E. Oja, Independent Component Analysis, John Wiley & Sons, New York, 2001
[14]Ioanna Christoyianni, “Computer aided diagnosis of breast cancer in digitized mammograms” Computerized Medical Imaging and Graphics, v 26, n 5, September/October, 2002, p 309-319
[15]Rafayah Mousa et al. “Breast cancer diagnosis system based on wavelet analysis and fuzzy-neural” Expert Systems with Applications, v 28, n 4, May, 2005, p 713-723
[16]Liyang Wei et al.“A study on several machine-learning methods for classification of malignant and benign clustered microcalcifications” IEEE Transactions on Volume 24, Issue 3, March 2005 Page(s):371-380
[17]Reyer Zwiggelaar et al. “ The benefit of knowing your linear structures in mammographic images” In Proceedings of Medical Image Understanding and Analysis ’02 , p 73-76, 2002.
[18]M. Sonka, V. Halavac & R. Boyle. Image Processing, Analysis and Machine Vision. Chapman and Hall Publishing, 1993.
[19]“數位像處理 活用Matlab” 繆紹綱編著 全華科技圖書股份有限公司
[20]“一個使用紋理特徵及羅吉斯回歸分辨超音波乳房腫塊良惡性的方法” 施泰宇著
[21]A. Hyvarinen, E. Oja “Independent component analysis: Algorithms and applications” Neural Networks, v 13, n 4, May, 2000, p 411-430
[22]M. S. Bartlett et al.“ Face recognition by independent component analysis” IEEE Transactions on Neural Networks, v 13, n 6, November, 2002, p 1450-1464
[23]Tee Connie et al. “Palmprint recognition with PCA and ICA” Proceedings of the 2003 ACM SIGMM workshop on Biometrics methods and applications, November 08, 2003, Berkley, California
[24]Jarmo Hurri et al.“Image feature extraction using independent component analysis” In Proc. NORSIG’96, Espoo, Finland
[25]V. Vapnik, Statistical Learning Theory. New York: Wiley,1998.
[26]Ted C. Wang, Nicolaos B. Karayiannis, “Detection of microcalcifications in digital mammograms using wavelets” IEEE Trans. Med. Img., vol. 17, no. 4, Aug. 1998.
[27]Shawe-Taylor, N.C.a.J., “An Introduction to Support Vector Machines and Other Kernel-based Learning Methods” 2000: Cambridge University Press.
[28]Y. Jiang, R. M. Nishikawa et al. “Improving breast cancer diagnosis with computer-aided diagnosis,” Academic Radiol., vol. 6, pp. 22-23, 1999.
[29]Sung-Nien Yu, Kuen-Yuei Li, Yu-Kun Huang,“Detection of microcalcification in digital mammograms using wavelet filter and Markov random field model” Computerized Medical Imaging and Graphics 30 (2006) 163-173
[30]“模式識別” 邊肇祺著 清華大學出版社
[31]“類神經網路” 羅華強編著 清蔚科技出版
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top