臺灣博碩士論文加值系統

心房纖維性顫動起因於竇房結產生的動作電位在心房肌肉上混亂傳導，部分活化心肌的電流反覆地再興奮原來的區域，持續地循環刺激導致心房某些部位的收縮步調無法一致而降低其輸血功能。電腦斷層掃描可以記錄心臟搏動過程的瞬間影像並以數位化儲存，藉由觀察多時序影像之間的左心房收放型態變化可以應用於分析異常節律性心房腔室壁之運動。
計算心房壁運動首先要克服不同時序影像中相同參考依據座標校正的問題，在型態變化的分析領域裡，圖形配準常被作為完成影像之間特徵參考圖形匹配之方法，本研究利用高品質的心臟電腦斷層影像，有效率地計算影像之間的匹配程度，排除造影過程中非左心房運動的干擾。在心房的輪廓圈選是使用區域成長演算法，利用這些輪廓的資料計算房壁運動來評估左心房在各區域的舒張及收縮功能。
透過本研究開發之系統工具分析二十組的心臟電腦斷層影像（11個正常人，9個心房纖維性顫動患者），結果顯示，正常人與患有心房纖維性顫動的病患，位於左心房的左下方前、後側靠近僧帽瓣處有顯著差異：正常人的前、後側平均移動量各為6.37±1.81mm及7.01±1.72mm；AF患者的前、後側平均移動量各為8.76±1.46mm及9.20±1.63mm，p value<0.01。此外，以向量分析環繞右下肺靜脈周圍的心房壁下、後區塊亦有顯著差異：正常人的下、後區域向量大小變化分別為0.045±0.016及0.051±0.022；AF患者的下、後區域向量大小變化各為0.089±0.038及0.085±0.028，p value<0.01。代表著患有心房纖維性顫動的病患會導致左心房局部功能異常，在臨床診斷與治療上便可藉由型態的改變作為判斷的指標。

Atrial fibrillation is caused by the disorderly action voltage which generates by sinoatrial node in the atrium. Some of the excited cardiac muscle patterns are re-stimulated by the recurrent impulse which makes the left atrium contract irregularly. As the heartbeat goes by, these cardiac patterns change can be taken down immediately by the CT scan. With the various time of CT images, we can analyze the abnormal rhythmic wall motion of heart chamber.
For calculation of wall motion, reference which being a fixed figure in different time of CT image series must be calibrated slice by slice. In the analytic field of pattern changes, image registration is often taken as the method to complete characteristic reference matches between images. This study utilizes the high-quality cardiac CT images to efficiently calculate the matching degree between the CT images so that we can rule out the interference occasioned by the non-left atrium motion during the CT scan. As to delineate the atrial profile, we adopt seed region growth algorithm. By means of the profile data, we can analyze the atrial wall motion to evaluate the contraction extent of all regions in the left atrium.
With the system built by this research, analysis of the 20 cardiac CT images (eleven being normal and nine suffering from atrial fibrillation) indicates the obvious differences between the normal people and the patients at issue. For the normal ones, the average side motion at anterior and posterior of left inferior wall readings are 6.37±1.81mm and 7.01±1.72mm; for the patients, the readings are 8.76±1.46mm and 9.20±1.63mm，p value<0.01. Besides, there is another difference with regard to the vector analysis of LA areas circling the right inferior pulmonary vein. For the normal, the magnitude of difference at inferior and posterior patterns readings are 0.045±0.016 and 0.051±0.022; and for the patients, the readings are 0.089±0.038 and 0.085±0.028，p value<0.01. These analyses implicate that the AF patients would develop partial malfunction in the left atrium. Therefore, they will serve as the diagnose index when clinic diagnoses and treatments are in process.

目錄
摘要 I
ABSTRACT II
謝誌 IV
目錄 V
圖索引 VIII
表索引 X
第一章 緒論 1
1-1 前言 1
1-2 研究背景 2
1-3 文獻探討 3
1-4 研究目的 5
1-5 論文架構 6
第二章 基本理論 7
2-1 左心房結構 7
2-1-1 生理結構 8
2-1-2 基礎病理 8
2-1-3 心跳週期 10
2-2 DICOM影像之簡述與解碼 11
2-2-1 DICOM檔案格式 12
2-2-2 DICOM檔案內容 13
2-3 邊緣檢測 14
2-4 幾何中心 16
2-5 三維模型重組與電腦繪圖 18
2-5-1 OpenGL 19
2-5-2 材質貼圖 21
2-5-3 三維物件與平面的旋轉和位移 22
2-6 重新取樣 23
2-7 參數分析 30
2-7-1 以Bull’s eye分析左心房運動 31
2-7-2 以向量分析左心房運動 32
2-8 本章總結 33
第三章 系統流程與架構 34
3-1 系統流程 34
3-2 圖形配準（IMAGE REGISTRATION） 36
3-2-1 鑑別（Identification） 36
3-2-2 匹配（Matching） 37
3-2-3 驗證（Verification） 40
3-3 影像重新取樣（IMAGE RESAMPLING） 41
3-4 牛眼圖分析心肌運動 45
3-5 向量分析心肌運動 46
3-5-1 肺靜脈與左心房截面的選取 47
3-5-2 劃分左心房 49
3-5-3 雷達圖 50
3-6 系統顯示 52
3-6-1 牛眼圖（Bull’s Eye） 52
3-6-2 雷達圖（Radar Plot） 53
3-7 系統檔案架構 56
3-7-1 DICOM檔案架構 57
3-7-2 標頭檔 59
3-7-3 肺靜脈資訊 60
3-7-4 Information 61
3-8 本章總結 61
第四章 實驗結果與討論 62
4-1 系統操作介面介紹 62
4-1-1 圖形鑑別介面 62
4-1-2 圖形比對介面 64
4-1-3 圖形驗證介面 66
4-1-4 重新取樣介面 69
4-1-5 肺靜脈圈選介面 70
4-2 系統結果分析 73
4-2-1 牛眼圖分析 73
4-2-2 向量分析 75
4-3 本章總結 76
第五章 結論與未來發展 77
5-1 結論 77
5-2 未來發展 78
參考文獻 79
附錄A CT影像匹配結果 83
附錄B 二十組向量分析結果 91
心房纖維性顫動病人： 91
正常受測者： 94
作者簡介 96


圖索引
圖 2.1 心臟解剖圖 9
圖 2.2 心跳週期 11
圖 2.3 CT影像時序 11
圖 2.4 DICOM檔案結構 12
圖 2.5 DICOM檔案內容 13
圖 2.6 區域成長搜尋 15
圖 2.7 區域成長特殊情況 15
圖 2.8 各種邊緣收斂方式 16
圖 2.9 OPENGL程式控制流程 20
圖 2.10 重新取樣示意圖 24
圖 2.11 牛眼圖 31
圖 3.1 系統流程圖 35
圖 3.2 單張電腦斷層切片 36
圖 3.3 圖示說明匹配之方法 38
圖 3.4 匹配流程圖 39
圖 3.5 驗証流程圖 40
圖 3.6 重新取樣流程圖 42
圖 3.7 三維陣列模型 43
圖 3.8 圖示重新取樣流程 44
圖 3.9 輻射狀重新取樣30點 45
圖 3.10 左心房表面法向量圖 46
圖 3.11 左心房劃分流程圖 47
圖 3.12 取得左心房邊緣情形 48
圖 3.13 用向量分析的左心房區域 48
圖 3.14 法向量之計算示意圖 50
圖 3.15 雷達圖 51
圖 3.16 雷達圖展開 52
圖 3.17 900區牛眼圖 53
圖 3.18 18區牛眼圖 53
圖 3.19 雷達圖顯示視窗 55
圖 3.20 雷達圖展開之顯示視窗 56
圖 3.21 系統檔案架構 57
圖 3.22 標頭檔格式 59
圖 3.23 肺靜脈切面儲存格式 60
圖 3.24 取樣資訊檔案格式 60
圖 3.25 INFORMATION檔案格式 61
圖 4.1 圈選胸骨操作介面 63
圖 4.2 圈選胸骨遭遇狀況 63
圖 4.3 比對顯示介面 64
圖 4.4 驗證顯示介面 66
圖 4.5 圖形配準所用的視窗 68
圖 4.6 取樣前的影像 68
圖 4.7取樣後的影像 68
圖 4.8 圈選左心房輪廓操作介面 69
圖 4.9 修正取樣影像流程 70
圖 4.10 心臟三維模型顯示介面 71
圖 4.11 肺靜脈截面影像 72
圖 4.12 靠近肺靜脈的左心房影像 72
圖 4.13 直方圖 74

表索引
表 3.1 資料元件基本架構 58
表 3.2 影像像素模組 59
表 4.1 其中一組SUBJECT的驗證結果 67
表 4.2 牛眼圖分析（18區）統計結果 74
表 4.3 直方圖統計表 74
表 4.4 向量分析-振幅表 75
表 4.5 向量分析-相位表 76

參考文獻
﹝1﹞Zhaohui Zhang and Veronique Prinet, “A Rough–to–fine Satellite Image Registration Method with Sub-pixel Accuracy,” IEEE International Conference on Image Processing, vol. 3, pp. 385-388, 2002.
﹝2﹞Everine B. van de Kraats, Graeme P. Penney, Theo van Walsum, and Wiro J. Niessen, “Standardized Evaluation Methodology for 2-D – 3-D Registration,” IEEE Transactions on Medical Imaging, vol. 24, pp. 1177-1189, 2005.
﹝3﹞H.–M. Tsao, W.–C. Yu, H.–C. Cheng, M.–H. Wu, C.–T. Tai, W.–S. Lin, Y.–A. Ding, M.–S. Chang, and S.–A. Chen, “Pulmonary Vein Dilation in Patients with Atrial Fibrillation：Detection by Magnetic Resonance Imaging,” Journal of Cardiovascular Electrophysiology, vol. 12, pp. 809-813, 2001.
﹝4﹞S. I. Choi, J. B. Seo, S. H. Choi, S.-H. Lee, K.-H. Do, S. M. Ko, J. S. Lee, J.-W. Song, K.-S. Song, K.-J. Choi, Y.-H. Kim, and T.-H. Lim, “Variation of the size of pulmonary venous ostia during the cardiac cycle: optimal reconstruction window at ECG-gated multi-detecter row CT,” Eur Radiol, vol. 15, pp. 1441-1445, 2005.
﹝5﹞Yun-Hyeon Kim, Edith M. Marom, James E. Herndon II, and H. Page McAdams, “Pulmonary Vein Diameter, Cross-sectional Area, and Shape: CT Analysis,” Radiology, vol. 235, pp. 43-49, 2005.
﹝6﹞T.-H. Hauser, S. B. Yeon, S. McClennen, G. Katsimaglis, K. V. Kissinger, M. E. Josephson, N. M. Rofsky, and W. J. Manning, “Subclinical pulmonary vein narrowing after ablation for atrial fibrillation,” Heart, vol. 91, pp. 672-673, 2005.

﹝7﹞Daw-Tung Lin, Chung-Chih Lei, and Siu-Wan Hung, “Computer-Aided Kidney Segmentation on Abdominal CT Images,” IEEE Transactions on Information Technology in Biomedicine, vol. 10, pp. 59-65, 2006.
﹝8﹞Yan Kang, Klaus Engelke, and A. Kalender, “A New Accurate and Precise 3-D Segmentation Method for Skeletal Structure in Volumetric CT Data,” IEEE Transactions on medical Imaging, vol. 22, pp. 586-598, 2003.
﹝9﹞R. Adams and L. Bischof, “Seeded Region Growing,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 16, pp. 641-647, 1994.
﹝10﹞Stephen M. Pizer, P. Thomas Fletcher, Sarang Joshi, Andrew Thall, James Z. Chen, Yonatan Fridman, Daniel S. Fritsch, and A. Graham Gash, “Deformable M-Reps for 3D Medical Image Segmentation,” International Journal of Computer Vision, vol. 55, pp. 85-106, 2003.
﹝11﹞Ling Xia, Meimei Huo, Qing Wei, Feng Liu and Stuart Crozier, “Analysis of Cardiac Ventricular Wall Motion Based on a Three-dimensional Electromechanical Biventricular Model,” Physics in Medicine and Biology, vol. 50, pp. 1901-1917, 2005.
﹝12﹞Hideki Yamamoto, Tetsuya Sano, Tsuneharu Morito and Yoshio Hiraki, “Image Analysis of Left Ventricle Wall Motion by MRI Tagging,” Technology and Applications, International Workshop, pp. 91-94, 2001.
﹝13﹞Pengcheng Shi, Albert J. Sinusas, R. Todd Constable, Erik Ritman, and James S. Duncan, “Point-Tracked Quantitative Analysis of Left Ventricular Surface Motion from 3-D Image Sequences,” IEEE Transactions on Medical Imaging, vol. 19, pp. 36-50, 2000.

﹝14﹞Arkadiusz Sitek, Gregory J. Klein, Bryan W. Reutter, Ronald H. Huesman, and Grant T. Gullberg, “Measurement of the Biomechanics of 3-D Cardiac Function with Gated Nuclear Medicine Studies,” IEEE Nuclear Science Symposium Conference Record, vol. 4, pp. 2346-2349, 2005.
﹝15﹞David Levin, Usaf Aladl, Guido Germano, and Piotr Slomka, “Techniques for Efficient, Real-Time, 3D Visualization of Multi-Modality Cardiac Data Using Consumer Graphics Hardware,” Computerized Medical Imaging and Graphics, vol. 29, pp. 463-475, 2005.
﹝16﹞Meei-Ling Jan, Keh-Shih Chuang, Guo-Wei, Yu-Ching Ni, Sharon Chen, Chih-Hsien Chang, “A Three-Dimensional Registration Method for Automated Fusion of Micro PET-CT-SPECT Whole-Body Images,” IEEE Transactions on Medical Imaging, vol. 24, pp. 886-893.
﹝17﹞George D. Stetten and Stephen M. Pizer, “Medial-Node Models to Identify and Measure Objects in Real-Time 3-D Echocardiography,” IEEE Transactions on Medical Imaging, vol. 18, pp. 1025-1034, 1999. Segmentation of thrombus in abdominal aortic aneurysms from CTA with nonparametric statistical grey level appearance modeling
﹝18﹞S. D. Olabarriaga, J.-M. Rouet, M. Fradkin, M. Breeuwer, W. J. Niessen, “Segmentation of Thrombus in Abdominal Aortic Aneurysms From CTA With Nonparametric Statistical Grey Level Appearance Modeling,” vol. 24, pp. 477-485, 2005.
﹝19﹞Oussama M. Wazni, Hsuan-Ming Tsao, Shih-Ann Chen, Hsuan-Hung Chuang, Andrea Natale, and Allan L. Klein, “Cardiovascular Imaging in the Management of Atrial Fibrillation,” Journal of the American College of Cardiology, vol. 48, pp. 2077-2084, 2006.

﹝20﹞Hsuan-Ming Tsao, Mei-Han Wu, Wen-Chung Yu, Ching-Tai Tai, Yung-Kuo Lin, Ming-Hsiung Hsieh, Yu-An Ding, Mau-Song Chang, and Shih-Ann Chen, “Role of Right Middle Pulmonary Vein in Patients with Paroxysmal Atrial Fibrillation,” Journal of Cardiovascular Electrophysiology, vol. 12, pp. 1353-1357, 2001.
﹝21﹞Tran Thong, James McNames, Mateo Aboy, and Brahm Goldstein, “Prediction of paroxysmal Atrial Fibrillation by Analysis of Atrial Premature Complexes,” IEEE Transactions on Biomedical Engineering, vol. 51, pp. 561-569, 2004.
﹝22﹞陳巨弦, “實時四維超音波動態心臟影像顯示及分析系統,” 中壢中原大學, 2005.
﹝23﹞吳明鴻, “利用心臟電腦斷層影像評估左心房與左心耳之射血功能,” 中壢中原大學, 2007.
﹝24﹞麥麗敏, 新編生理學: 永大書局, 1999.
﹝25﹞ARTHUR C. GUYTON, “Human Physiology and Mechanisms of Disease,” 華杏出版股份有限公司, 1994.
﹝26﹞光復書局編輯部編著(黃蓮奇譯), 心臟和循環系統: 光復書局, 1996.
﹝27﹞繆紹綱, 數位影像處理: 全華, 1999.
﹝28﹞黃俊英, 多變量分析: 中國經濟企業研究所, 1991.
﹝29﹞劉霈, 線性代數: 滄海書局, 2004.
﹝30﹞蔡孟儒, “OpenGL在數學動畫上的應用,” 台中靜宜大學, 2004.