臺灣博碩士論文加值系統

醫學影像對於診斷上一直有很大的幫助，在醫學知識日新月異下，自動化的判斷一直是工程人員最終的目標，癲癇就是一種常見的腦部病變症狀，通常是腦部某些區塊病變或是腦部曾經有外傷或重擊，造成腦部細胞突然異常的過度放電所引發的腦部功能失調，本研究是希望能將病人貼上癲癇電極之後的CT影像空間座標給截取出來，為的是希望得到如此的空間座標後，能夠結合電極電壓做電壓運算，藉以求出腦部異常區域大小和區塊位置，由於癲癇病人的CT影像中含有大量腦組織的影像成分，因此利用演算法濾除腦組織與電線儀器等影像成分進而求取影像中只包含電極的座標成了本研究所著重的項目，之後便可以減少醫療人員需要使用套裝軟體手動標訂出空間電極座標所需的時間，以節省人力資源。

Epilepsy is one kind of common brain pathological symptom, usually due to certain sub-area pathological change , or brain injury , etc.
In this study , an algorithm for identifying electrodes from brain CT image proposed , in order to incorporate electrode information into brain image .
The major work is to find the locations of electrodes from brain CT images which are composed of massive brain tissues , bone electrical lines , and electrodes .
The proposed algorithm can reduce the efforts and errors in manual identification of electrodes.

目 錄
第一章　緒論............................................. 01
　　1.1　前言.............................................01
　　1.2　研究動機與目的...................................02
　　1.3　論文架構.........................................04
第二章 CT 灰階影像的特性與處理.......................... 05
　　2-1　癲癇疾病介紹.....................................05
　　2-2　處理CT成像......................................07
　　2-3　介紹Histogram原理...............................11
第三章 建構輸出點與找尋電極64個空間座標點...............18
3-1　整理架構流程.....................................18
3-2　求出每張CT影像上的物件中心點....................19
3-3　介紹K-mean分類演算法............................23
3-4　利用鄰近點搜尋方式將資料點作分類.................28
3-5　電極貼片座標特性與點數處理.......................36
第四章　結果與分析....................................... 44
　　4-1　處理之後顯示結果................................ 44
4-2 利用其他方式讀取影像資料........................ 47
　　4-3　將電位訊號和電極座標做Mapping...................51
第五章　未來展望.........................................52
參考文獻.................................................53





圖目錄
圖2-1　輸出點數程式流程圖................................07
圖2-2　Avizo軟體圖.......................................08
圖2-3　重疊組合示意圖....................................09
圖2-4-1..................................................12
圖2-4-2..................................................12
圖2-4-3..................................................13
圖2-5　經過Threshold之後的影像資料做Isosurface結果......14
圖2-6　將資料轉換成網格之後的顯示........................15
圖2-7　資料輸出結果......................................16
圖2-8　將輸出點數代回Avizo之後的驗證圖形................17
圖2-9　Avizo將影像轉成點座標的過程與方法.................17
圖3-1....................................................18
圖3-2-1　尋找每個物件的中心座標示意圖.................... 20
圖3-2-2..................................................21
圖3-3-1..................................................22
圖3-3-2　將每個物件的中心算出的簡單示意圖................22
圖3-4....................................................27
圖3-5....................................................28
圖3-6....................................................30
圖3-7-1..................................................31
圖3-7-2　經由改良k-mean之後跑出的結果...................32
圖3-8....................................................33
圖3-9-1..................................................34
圖3-9-2..................................................34
圖3-10...................................................35
圖3-11...................................................37
圖3-12...................................................38
圖3-13...................................................39
圖3-14...................................................40
圖3-15...................................................40
圖3-16...................................................41
圖3-17...................................................42
圖3-18...................................................43
圖4-1....................................................44
圖4-2....................................................45
圖4-3....................................................45
圖4-4....................................................46
圖4-5....................................................47
圖4-6　Labview程式架構圖.................................48
圖4-7-1　處理過後的單張影像，只保留了電極資訊.............48
圖4-7-2　利用4向的搜尋方向將物件群組中心算出............49
圖4-8....................................................49
圖4-9....................................................51

參考文獻

[1] H. Xiong, J. Wu, and J. Chen, “K-means clustering versus validation
measures: A data distribution perspective,” in Proc. 12th ACM SIGKDD
Int. Conf. Knowl. Discov. Data Mining, 2006, pp. 779–784.

[2] P.-N. Tan, M. Steinbach, and V. Kumar, Introduction to Data Mining.
Reading, MA: Addison-Wesley, 2005.

[3] K. Krishna and M. Narasimha Murty, “Genetic K-means algorithm,”
IEEE Trans. Syst., Man, Cybern. B, Cybern., vol. 29, no. 3, pp. 433–439,
Jun. 1999.

[4] T.S. Chen, Y.T. Chen, C.C. Lin, and R.C. Chen, “A Combined K-Means and Hierarchical Clustering Method For Improving the Clustering Efficiency of Microarrary,” In Proceedings of 2005 International Symposium on Intelligent Signal Processing and Communications Systems, 2005, pp. 405-408.

[5] Bookstein FL. Morphometrics. In: Toga AW, editor. Three-dimensional neuroimaging.
New York: Raven Press; 1990 [Chapter 8].

[6]Butson CR, Maks CB, McIntyre CC. Sources and effects of electrode impedance during deep brain stimulation. Clin Neurophysiol 2006;117(2):447–54.

[7] Butson CR, McIntyre CC. Tissue and electrode capacitance reduce neural activation volumes during deep brain stimulation. Clin Neurophysiol
2005;116(10):2490–500.

[8] Butson CR, McIntyre CC. Role of electrode design on the volume of tissue
activated during deep brain stimulation. J Neural Eng 2006;3(1):1–8.

[9] Hastreiter P, Rezk-Salama C, Soza G, Bauer M, Greiner G, Fahlbusch R, et al.
Strategies for brain shift evaluation. Med Image Anal 2004;8(4):447–64.