臺灣博碩士論文加值系統

醫療用電腦輔助診斷系統(CAD, computer-aided detection)目前已廣泛應用在臨床醫學診斷上。隨著科技的進步醫學診斷與電腦相應結合，目前成像儀器有斷層掃瞄(CT)、磁振造影(MRI)、超音波(Ultrasound)等。此儀器影像藉由網路(PACS)可傳輸至電腦端進行影像處理，電腦端可將輸入之影像進行、分割、偵測和影像重建等處理，醫師藉由這些輔助可立即發現病灶，也大幅減少判讀時間。MRI(Magnetic Resonance Imaging)是近年來在臨床診斷上相當重要的影像工具，此種使用準確而不必侵入人體就可以得到人體內部器官造影的方法，對醫學的診斷、醫療和後續工作都十分重要。

在目前醫學上面發現，心血管疾病是最常見的死因。所以在臨床或者醫學上，心臟的結構和功能是非常重要的。在目前診斷上面，主要針對心臟功能的評估是藉由左心室(Left Ventricular)和右心室(Right Ventricular)在舒張末期(end diastolic)和收縮末期(end systolic)進行評估。為了在臨床實踐上，MRI資料必須被充分的利用，並且考慮資料的總數量來自動對心室做評估。在過去的幾年裡，已經有人提出針對左心室(Left Ventricular)使用半自動和全自動分割的技術。因為，左心室(Left Ventricular)分割一直是一個很特別的重點，在於左心室(Left Ventricular)比右心室(Right Ventricular)面積來的大、強壯和有規則性，並且在分割執行上面左心室(Left Ventricular)比右心室(Right Ventricular)來的容易多。所以，在目前現有的文獻方面，針對心臟結構的評估幾乎都是藉由左心室(Left Ventricular)分割的技術來探討。不過近年來，發現磁振造影(MRI，Magnetic Resonance Imaging) 對右心室(Right Ventricular)影像有很高的分辨率，所以近年來也越來越多人使用右心室(Right Ventricular)評估來當標準工具。不過將心臟使用在磁振造影(MRI，Magnetic Resonance Imaging)針對右心室(Right Ventricular)分割上是有很大的困難，因為右心室(Right Ventricular)有複雜的運動和分析，所以伴隨著許多的變數以及形狀屬於月牙形狀和心室壁很薄等，以上總總說明了右心室的分析和分割上的複雜度和困難。

本研究設計了一套演算法是針對四維胸腔MRI(Magnetic Resonance imaging)進行分割與重建。我們提供全自動分割方式解決手動分割方式所帶來的問題(耗時長、人力需求大)和心臟使用在磁振造影(MRI，Magnetic Resonance Imaging)針對右心室(Right Ventricular)分割上帶來的困惱。本研究設計一個技術與現有文獻上技術的不同點是在於，本研究的技術是用心臟運動模式的概念，應用在左心室(Left Ventricular)和右心室(Right Ventricular) 全自動分割上 。

Nowadays, computer-aided detection has been widely used in clinical diagnosis. With advances in technology combined with medical diagnosis and computer, Imaging has tomography imaging equipment (CT), magnetic resonance imaging (MRI), ultrasound (Ultrasound), etc. Through network (PACS) this instrument image can be transferred to PC for image processing. Computer terminal can operate, segment, detect and image reconstruct entered images. By these auxiliary tools, physicians can immediately discovered lesions and significantly reduce diagnosis time. MRI has been an important imaging tools for clinical diagnosis in recent years. It is important method for clinical diagnosis, medical treatment and subsequence that MRI can obtain internal organs imaging accurately without invading human body.

In the current medical findings, cardiovascular disease is the most common cause of death. Therefore, cardiac structure and function are very important in the clinical. The current diagnosis show that the main assessment aimed at cardiac function is through end-diastole and end- diastolic of left ventricular and right ventricular. In order to implement in clinical diagnosis, MRI needs to be use sufficiently and consider total data to assess ventricular automatically.In the past few years, study has proposed using semi-automatic and fully automatic segmentation technique for the left ventricle. Left ventricular has been a particular point because left ventricle’s area is larger, stronger and more regular than right ventricle’s. Moreover, it is easier to segment on left ventricular than right ventricular. Therefore, in the existing literature, most studies within cardiac structure focus on left ventricular evaluation. However, in recent years, researchers found that MRI has high resolution on the right ventricle images, so there are more and more people use the right ventricle assessments as a standard tool. Nevertheless, it is extremely difficult using MRI on right ventricular segmentation, because the right ventricle has a complex motion and analysis, following many variables the shape is crescent and ventricular walls are quite thin. To sum up, it is complex and difficult analyzing and segmenting on right ventricular.

In this study, an algorithm is designed for four-dimensional chest MRI segmentation and reconstruction. We offer a fully automated segmentation method to resolve the issue which manual segmentation method results in (e.g., wastes time and demands a lot of human resources) and the problem that using MRI on right ventricular. The significantly technical difference between previous studies, this study use the concept of cardiac motion mode and apply automatic segmentation to both left ventricular and right ventricular.

圖目錄
表目錄
第一章 緒論
1.1研究動機
1.2研究目標
1.3論文貢獻
1.4論文架構
第二章 研究背景
2.1左心室技術探討
2.2右心室技術探討
第三章 研究數據和方法
3.1研究數據
3.2研究方法
3.2.1前置處理
3.2.2左心室(Left Ventricular)的偵測
3.2.2.1簡單(Coarse)的左心室偵測應用在時間軸(T軸)
3.2.2.2準確(Precise)左心室偵測應用在Z軸
3.2.3右心室(Right Ventricular)的偵測
3.2.3.1右心室(Right Ventricular)的處理
3.2.3.1.1右心室(Right Ventricular)合併的處理
3.2.4對右心室(Right Ventricular)效正
3.2.4.1時間軸空間上對右心室進行效正
3.2.4.1.1時間軸空間上對右心室進行前效正
3.2.4.1.2時間軸空間上對右心室進行中間效正
3.2.4.1.3時間軸空間上對右心室進行後效正
3.2.4.2 Z軸空間上對右心室(Right Ventricular)進行效正
3.2.4.2.1 Z軸空間向上對右心室進行效正
3.2.4.2.2 Z軸空間向下對右心室進行效正
第四章 實驗方法和結果
4.1實驗方法
4.2實驗結果
4.2.1心內膜分割比較
4.2.2心外膜分割比較
第五章 結論與未來展望
5.1結論
5.2未來展望
參考文獻

[1]Y. Keller and A. Averbuch, "Multisensor image registration via implicit similarity," Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. 28, pp. 794-801, 2006.
[2]核磁共振(MRI)介紹, "http://zito1203.pixnet.net/blog/post/36666925-%E6%A0%B8%E7%A3%81%E5%85%B1%E6%8C%AF%28mri%29%E4%BB%8B%E7%B4%B9.."存取日期20130612
[3]秀傳醫院, "http://www.chimei.org.tw/ePhotoAlbum/files/6E3C220AC0E412C4BB6C5C7C89625971.pdf.." 存取日期20130612
[4]J. Caudron, J. Fares, P. H. Vivier, V. Lefebvre, C. Petitjean, and J. N. Dacher, "Diagnostic accuracy and variability of three semi-quantitative methods for assessing right ventricular systolic function from cardiac MRI in patients with acquired heart disease," Eur Radiol, vol. 21, pp. 2111-20, Oct 2011.
[5]F. Haddad, S. A. Hunt, D. N. Rosenthal, and D. J. Murphy, "Right ventricular function in cardiovascular disease, part I: Anatomy, physiology, aging, and functional assessment of the right ventricle," Circulation, vol. 117, pp. 1436-48, Mar 18 2008.
[6]A. Attili, A. Schuster, E. Nagel, J. Reiber, and R. van der Geest, "Quanti_cation in cardiac MRI: advances in image acquisition and processing," 2010.
[7]MICCAI. (2012). RV Segmentation Challenge in Cardiac MRI. . Available: http://www.litislab.eu/rvsc/information-for-participants/rv-segmentation- challenge-in-cardiac-mri.
[8]M. Sermesant, C. Forest, X. Pennec, H. Delingette, and N. Ayache, "Deformable biomechanical models: Application to 4d cardiac image analysis.," Medical Image Analysis, vol. 7, pp. 475-488, 2003.
[9]S. C. Mitchell, J. G. Bosch, B. P. F. Lelieveldt, R. J. van der Geest, J. H. C. Reiber, and M. Sonka, "3-D active appearance models: segmentation of cardiac MR and ultrasound images," Medical Imaging, IEEE Transactions on, vol. 21, pp. 1167-1178, 2002.
[10]N. Paragios, "A Variational Approach for the Segmentation of the Left Ventricle in Cardiac Image Analysis," Int J Comput Vision, vol. 50, pp. 345-362, 2002.
[11]H. C. van Assen, M. G. Danilouchkine, M. S. Dirksen, J. C. Reiber, and B. F. Lelieveldt, "A 3-D active shape model driven by fuzzy inference: Application to cardiac CT and MR," vol. 12, pp. 595-605, 2008.
[12]M. P. olly, H. Xue, L. Grady, and J. Guehring, "Combining registration and minimum surfaces for the segmentation of the left ventricle in cardiac cine MR images," Med Image Comput Comput Assist Interv, vol. 12, pp. 910-918, 2009.
[13]M. Lorenzo-Valdés, G. I. Sanchez-Ortiz, A. G. Elkington, R. H. Mohiaddin, and D. Rueckert, " Segmentation of 4D cardiac MR images using a probabilistic atlas and the EM algorithm," Med Image Anal, vol. 8, pp. 255-265, 2004.
[14]M. Fahmi Khalifa, IEEE., G. M. Beache, G. Gimel'farb, G. A.Giridharan., and M. Ayman El-Baz*, IEEE., "Accurate Automatic Analysis of Cardiac Cine Images," TRANSACTIONS ON BIOMEDICAL ENGINEERING, vol. 59, 2012.
[15]J. Peters, O. Ecabert, C. Meyer, R. Kneser, and J. Weese, "Optimizing boundary detection via simulated search with applications to multi-modal heart segmentation. ," Med Image Anal, vol. 14, pp. 70-84, 2010.
[16]X. Zhuang, K. Rhode, S. Arridge, R. Razavi, D. Hill, D. Hawkes, et al., "An atlas-based segmentation propagation framework locally a_ne registration_ application to automatic whole heart segmentation.," Med Image Comput Comput Assist Interv, vol. 11, pp. 425-433, 2008.
[17]M. Hullebrand., A. Hennemuth., D. Messroghli., T. Kuhne., O. Friman., B. G. Fraunhofer MEVIS - Institute for Medical Image Computing, et al., "Semi-Automatic 4D Fuzzy Connectedness Segmentation of Heart Ventricles in Cine MRI."
[18]Z. Yun, X. Papademetris, A. J. Sinusas, and J. S. Duncan, "Segmentation of the Left Ventricle From Cardiac MR Images Using a Subject-Specific Dynamical Model," Medical Imaging, IEEE Transactions on, vol. 29, pp. 669-687, 2010.
[19]W. Xiaoxu, H. Qingmao, and H. Pheng Ann, "Left ventricle segmentation with mixture of Gaussian active contours," in Biomedical Imaging (ISBI), 2012 9th IEEE International Symposium on, 2012, pp. 230-233.
[20]L. D. Cohen, "On active contour mod," Computer Vision,Graphics, and Image Processing, vol. 53, pp. 211-218, 1989.
[21]H. Xiaolei and D. N. Metaxas, "Metamorphs: Deformable Shape and Appearance Models," Pattern Analysis and Machine Intelligence, IEEE Transactions on, vol. 30, pp. 1444-1459, 2008.
[22]W. E. Lorensen and H. E. Cline, "Marching cubes: A high resolution 3d surface construction algorithm," Computer Graphics, vol. 21, pp. 163-169, 1987.
[23]R.Sreemathy., Rekha.S.Patil., P. Pune Institute of Computer Technology, India, and K. Dr.D.Y.Patil Technical Campus , India, "Computer assisted analysis of left ventricle using level set and enhanced filtering in cardiac MRI," 2012.
[24]C. L., C. X., C. G., and M. D. F., "Level set evolution without re-initialization: a new variational formulation," IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2005.
[25]R. E. Berbari, N. Kachenoura, A. Redheuil, A. Herment, E. M. I Bloch3, F. Frouin, et al., "An Automated Evaluation of Regional Left Ventricular Function on Cine Magnetic Resonance Images," Computers in Cardiology, vol. 35, pp. 777-780, 2008.
[26]C. M, W. N, D. V, J. A, K. S, L. W, et al., "Standardized myocardial segmentation and nomenclature for tomographic imaging of the heart: a statement for healthcare professionals from the cardiac imaging committee of the council on clinical cardiology of the american heart association.," vol. 105, pp. 539-542, 2002.
[27]E. B. R, F. F, R. A, A. E, M. E, B. I, et al., "D′eveloppement et ′evaluation d′une m′ethode de segmentation automatique de l′endocarde sur des images acquises par r′esonance magn′etique.," ITBMRBM, vol. 28, pp. 117-123, 2007.
[28]M. R. Kaus, J. von Berg, J. Weese, W. Niessen, and V. Pekar, "Automated segmentation of the left ventricle in cardiac MRI," Med Image Anal, vol. 8, pp. 245-54, Sep 2004.
[29]C. Pluempitiwiriyawej, J. Moura, W. Yi-Jen Lin, and H. Chien, "Cardiac MR image segmentation: quality assessment of STACS," in Biomedical Imaging: Nano to Macro, 2004. IEEE International Symposium on, 2004, pp. 828-831 Vol. 1.
[30]C. Cocosco, W. N. Wiro, T. Netsch, E.-J. Vonken, G. Lund, A. Stork, et al., "Automatic image-driven segmentation of the ventricles in cardiac cine MRI," J Magn Reson Imaging, vol. 28, pp. 366-74, 2008.
[31]S. C. Mitchell, B. P. F. Lelieveldt, R. J. van der Geest, H. G. Bosch, J. H. C. Reiver, and M. Sonka, "Multistage hybrid active appearance model matching: segmentation of left and right ventricles in cardiac MR images," Medical Imaging, IEEE Transactions on, vol. 20, pp. 415-423, 2001.
[32]S. Ordas, L. Boisrobert, M. Huguet, and A. F. Frangi, "Active shape models with invariant optimal features (IOF-ASM) application to cardiac MRI segmentation," in Computers in Cardiology, 2003, 2003, pp. 633-636.
[33]J. Caudron, J. Fares, V. Lefebvre, P.-H. Vivier, C. Petitjean, and J.-N. Dacher, "Cardiac MR assessment of right ventricular function in acquired heart disease: Factors of variability," Academic Radiology, vol. 19, pp. 991-1002, 2012.
[34]A. Tsai, A. Yezzi, Jr., W. Wells, C. Tempany, D. Tucker, A. Fan, et al., "A shape-based approach to the segmentation of medical imagery using level sets," Medical Imaging, IEEE Transactions on, vol. 22, pp. 137-154, 2003.
[35]Y. Boykov and M. Jolly, "Interactive graph cuts for optimal boundary and region segmentation of objects in n-d images," International Conference on Computer Vision, vol. 1, pp. 303-308, 2001.
[36]Y. Li, J. Sun, C.-K. Tang, and H.-Y. Shum, " Lazy snapping. ACM Trans. Graph," vol. 1, pp. 105-113, 2004.
[37]J. Stawiaski, E. Decenciere, and F. Bidault, "Interactive liver tumor segmentation using graph-cuts and watershed.," Grand Challenge Liver Tumor Segmentation, 2008.
[38]M. Zuluaga, M. Cardoso, M. Modat, and S. Ourselin, "Multi-atlas propagation whole heart segmentation from mri and cta using a local normalised correlation coe_cient criterion.," vol. 7945, pp. 172-180, 2013.
[39]L. Bonnemains, D. Mandry, P. Y. Marie, E. Micard, B. Chen, and P. A. Vuissoz, "Assessment of right ventricle volumes and function by cardiac MRI: quantification of the regional and global interobserver variability," Magn Reson Med, vol. 67, pp. 1740-6, Jun 2012.
[40]D. Huttenlocher, G. Klanderman, and W. Rucklidge, "Comparing images using the hausdor distance," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 15, pp. 850-863, 1993.