臺灣博碩士論文加值系統

　　在醫學研究中，細胞的觀察是重要的基本程序，藉由細胞之特性去分析細胞是否處於異常狀態。在一般的顯微鏡底下，細胞幾乎是透明看不見的。有一種Phase Contrast的成像技術，其原理是利用物體間不同的折射率，會得到較佳對比效果。因為由一般肉眼的觀察及測量需花費很大的時間與精力，我們希望藉由影像處理發展一套自動追蹤組織細胞之系統協助醫學專家仔細評估細胞顯微影像，以找出各種病理現象或其成因。

　　本篇論文主要是對於受傷之上皮組織細胞，進行細胞分割、追蹤及分析細胞之特性。欲追蹤細胞在影像序列中的連續變化，可以藉由整張影像分割的方式或是局部尋找對應關係，找出細胞在對應影像中的邊緣。在本論文中，首先經由前處理改善影像品質，使得非分裂細胞內部與光暈差異性變大以利隨後細胞分割與追蹤。影像前處理後，細胞與背景之間之差異亦變大，即可由時間差異值影像分割出背景。利用canny邊緣偵測及分水嶺分割方法交叉比對，定義出初始影像中所要追蹤之初始細胞。依據細胞不同狀態之特性判斷出細胞主要之三個狀態: 間期、有絲分裂及細胞質分裂。我們依不同之狀態，利用適當之方法追蹤出細胞邊緣。當細胞屬於間期狀態時，追蹤方法主要為canny邊緣偵測、分水嶺分割、可變動性區塊比對和動態輪廓模型。若細胞屬於有絲分裂狀態則追蹤方法為限制圓形之動態輪廓模型。若細胞正進行細胞質分裂須先由動態輪廓搜尋分裂完成之概略區域，然後將分水嶺的分割結果若位於此概略區域內，其命名為候選子細胞區域。利用逆向追蹤確認候選子細胞區域是否為細胞分裂完成之子細胞。在細胞分析方面，依據細胞追蹤之結果有電腦自動追蹤和專家比較面積和中心之誤差、細胞三種狀態之分布比例、總輪廓數與人工手動調整的輪廓數之比例、還有由中心可以計算細胞之移動速率及方向。此外，觀察整體細胞的運動趨勢也是所要探討的問題之一。

　The observation of cells is one of the essential processes in fundamental medical study. Based on the observed properties, researchers can evaluate the conditions of cells. Cells are almost transparent and can not be seen under normal microscopes. Therefore, phase contrast imaging is adopted in these experiments for obtaining better contrast in cell images.

　However, it takes a lot of time and efforts to observe cells by human eyes. Therefore, it is certainly necessary to develop a computer system for tracking cells automatically. It can not only help pathologists for observing cell motion from a sequence of microscopic images but evaluating all kinds of pathology phenomenon or parameters of cell.

　This thesis is focused on cell segmentation, cell tracking, and analyzing the property of cells in the wounded area. For analyzing the continuous variation of cells in a sequence of images, it is important to correctly define the boundary of cells in the corresponding images. The cell boundaries are tracked, in the image sequence, based on either the global segmentation results or the matched local properties. In this thesis, we first improve the quality of image by image preprocessing, which enhances the intensity variations between interior of cells and halo. This preprocessing step can facilitate the subsequent cell tracking. After image preprocessing, the background was removed based on the temporal difference image. Then, we compared the segmented regions generated from canny edge detector and watershed algorithm to define the initial contours for cell tracking.

　To more accurately track a cell, cells are defined with three differente phases, which are interphase, mitotic, and cytokinesis phase. All of them are with different image properties that can be used to discriminate cell phases from the images. The methods for tracking interphase cells include canny edge detector, watershed segmentation, deformable block matching, and active contour model. Mitotic cells are tracked by active contour model with round shape constraint. As for cytokinesis cells, dividing region is first searched by using active contour model. Then, backward tracking is adopted to check the results of segmentation in the dividing region to decide whether it is the corresponding daughter cell.

　In the experiments, we compare the errors of area and center position calculated by the proposed system with respect to the manual tracking results by experts. Also, ratio of manually adjusted boundaries to total detected boundaries is calculated to evaluate the tracking robustness. From the experimental results, the proposed algorithm is found as accurate as manual tracking and is hence helpful for studing cell dynamics.

第一章、序論 1
1.1 研究動機及目的 1
1.2 相關研究 2
1.3 章節提要 4
第二章、細胞影像之前處理 5
2.1 細胞影像之強化 5
2.1.1 Otsu’s threshold 分類 5
2.1.2 混合高斯模型之分類 12
2.1.3 對比度之強化 16
2.2 背景分割 18
2.2.1 時間上差異值影像 19
2.2.2 背景邊界描繪及邊界修改 20
第三章、細胞分割及設定輪廓 24
3.1 整體細胞影像分割方法 24
3.1.1 Canny edge 分割方法 24
3.1.2 分水嶺分割方法 28
3.2 設定欲追蹤之細胞輪廓 34
第四章、細胞分析與追蹤 39
4.1細胞狀態分析 40
4.1.1 細胞影像結構及生命週期 40
4.1.2 細胞狀態之判別 42
4.2 間期狀態之追蹤 47
4.2.1 canny edge 及分水嶺分割結果之選取 48
4.2.2 簡介局部可變動性之區塊對應與動態輪廓模型 50
4.2.3 動態輪廓模型結合可變動的區塊比對 55
4.2.4 整體可變動性之區塊比對 60
4.2.5 重疊區域之互斥力動態輪廓 62
4.3 分裂期細胞追蹤 64
4.3.1 有絲分裂狀態追蹤 64
4.3.2 細胞質分裂狀態追蹤 67
第五章、實驗結果與討論 73
5.1系統概述 73
5.2 細胞序列影像的追蹤結果 74
5.3 細胞運動變化分析 79
5.4 實驗結果在3-D空間上的呈現 91
第六章、結論及未來展望 92
6.1 結論 92
6.2 未來展望 94
參考文獻 95
附錄 99

[ 1 ] Jim H. Velduis, G. Wayne Brodland , “A deformable block-matching algorithm for tracking epithelial cells” , Image and Vision Computing, vol. 17 ,pp. 905–911, 1999
[ 2 ] D. J. Williams and M. Shan, “A Fast Algorithm for Contours and Curvature Estimation”, CVGIP: Image Understanding, Vol. 55, No. 1, pp.14-26, 1992.
[ 3 ] Steven Lobregt and Max A. Viergever, “A Discrete Dynamic Contour Model”, IEEE Trans. on Medical Imaging, vol. 14, no. 1, pp.12-24, March 1995.
[ 4 ] http://life.nthu.edu.tw/~b831608/ps/cells2.html
[ 5 ] P. Arhets, J.-C. Olivo, P. Gounon, P. Sansonetti, and N. Guillèn, “Virulence
and functions of myosin II are inhibited by overexpression of light meromyosin in Entamoeba histolytica,” Mol. Biol. Cell., vol. 8, pp. 1537–1547, 1998.
[ 6 ]“Digital Image Processing” second edition, Rafael C. Gonzalez, Richard E.Woods , Prentice Hall, New Jersey, 2002
[ 7 ] P.S. UMESH ADIGA ,“Intergrated Approach for Segmentation of 3-D Confocal Image of a Tissue Specimen”, Microscopy Research and Technique vol.54, pp.267-270, 2001.
[ 8 ]“Image Processing , Analysis, and Machine Vision” second edition, Milan Sonka, Vaclav Hlavac, Roger Boyle , PWS , 1998
[ 9 ] Cheng-Hung Chuang and Wen-Nung Lie , “A Downstream Algorithm Based on Extended Gradient Vector Flow Field for Object Segmentation,” IEEE Transactions on Image Processing , vol.13, no.10, pp.1379-1392, October , 2004
[ 10 ] K. Wu, D. Gauthier, and M. D. Levine, “Live cell image segmentation,” IEEE Transactions on Biomedical Engineering, vol. 42, no. 1, pp.1-12, Jan. 1995.
[ 11 ] M. Kass, A. Witkin, and D. Terzopoulos, “Snakes: Active Contour Models,” International Journal of Computer Vision, vol. 1, no. 4, pp.321-331, 1987.
[ 12 ] T.F. Cootes , C.J. Taylor , D.H.Cooper , and J.Graham , “Active Shape Models －Their Training and Application,” Computer Vision and Image Understanding vol.61 , no.1 , pp.38-59, January 1995
[ 13 ] S. Sun, D. R. Haynor, and Y. Kim, ”Semiautomatic video object segmentation using vsnakes,” IEEE Transactions on Circuits Syst. Video Technol., vol.13, no.1, pp.75-82, Jan. 2003
[ 14 ] C. Xu, and J. L. Prince, “Snakes, shapes, and gradient vector flow,” IEEE Transactions on Image Processing, vol. 7, no. 3, pp.359-369, March 1998.
[ 15 ] W. P. Choi, K. M. Lam and W. C. Siu, “An adaptive active contour model for highly irregular boundaries,” Pattern Recognition, vol. 34, pp. 323-331, 2001.
[ 16 ] Y. Fu, A. T. Erdem, and A. M. Tekalp, “Tracking visible boundary of objects using occlusion adaptive motion snake," IEEE Transactions on Image Processing, vol. 9, no. 12, pp. 2051-2060, Dec. 2000.
[ 17 ] Chenyang Xu , Jerry L. Prince“Generalized gradient vector flow external forces for active contours,” Signal Processing , vol.71, pp.131-139, 1998
[ 18 ] Jinshan Tang, Scott T. Acton , “Vessel Boundary Tracking for Intravital Microscopy Via Multiscale Gradient Vector Flow Snakes,” IEEE Transactions on Biomedical Engineering , vol.51 , no.2, pp.361-324, Feb. 2004
[ 19 ] Y. L. Fok, J. C. K. Chan, and R. T. Chin, “Automated analysis of nerve-cell images using active contour models,” IEEE Transactions on Medical Imaging, vol. 15, no. 3, pp.353-368, Jun. 1996.
[ 20 ] D. P. Mukherjee, N. Ray, and S. T. Acton, “Level set analysis for leukocyte detection and tracking,” IEEE Transactions on Image Processing, vol. 13, no. 4, pp.562-572, Apr. 2004
[ 21 ] N. Ray, S. T. Acton, and K. Ley, “Tracking leukocytes in vivo with shape and size constrained active contours,” IEEE Transactions on Medical Imaging, vol. 21, no. 10, pp.1222-1235, Oct. 2002.
[ 22 ] N.Ray , S.T. Acton , “Motion Gradient Vector Flow : An External Force for Tracking Rolling Leukocytes With Shape and Size Constrained Active Contours” , IEEE Transactions on Medical Imaging , vol.23 , no.12, pp.1466-1478, Dec. 2004
[ 23 ] C. Zimmer, E. Labruyère, V. Meas-Yedid, N. Guillén, and J.-C. Olivo-Marin, “Segmentation and tracking of migrating cells in videomicroscopy with parametric active contours: a tool for cell-based drug testing,” IEEE Transactions on Medical Imaging, vol. 21, no. 10, pp.1212-1221, Oct. 2002.
[ 24 ] P.S. Umesh Adiga and B.B. Chaudhuri , “Efficient Cell Segmentation Tool for Confocal Microscopy Tissue Image and Quantitative Evaluation of FISH Signals “ , Microscopy Research and Technique vol.44, pp.49-68, 1999
[ 25 ] N.Eveno , A.caplier , and P.Coulon , “Accurate and Quasi-Automatic Lip Tracking” , IEEE Transactions on Circuits and Systems for Video Technology , vol.14 , no.5, pp.706-715, May. 2004
[ 26 ] A. Gyaourova, C. Kamath, S.-C. Cheung , “Block Matching for Object
Tracking”, UCRL-TR-200271, Oct. 2003
[ 27 ] O. Debeir , P. Van Ham , R. Kiss , C. Decaestecker , “Tracking of Migrating Cells Under Phase-Contrast Video Microscopy With Combined Mean-Shift Processes,” IEEE Transactions on Medical Imaging , vol.24 , no.6, pp.697-711, June 2005
[ 28 ] F. Germain, A. Doisy, X. Ronot, and P. Tracqui, “Characterization of cell deformation and migration using a parametric estimation of image motion,” IEEE Transactions on Biomedical Engineering, vol. 46, no. 5, pp584-600, May 1999.
[ 29 ] H. Hiral, R. Umegaki, M. Kino-Oka, and M. Taya, “Characterization of cellular motions through direct observation of individual cells at early stage,” Journal of Bioscience and Bioengineering, vol. 94, no. 4, pp.351-356, 2002.
[ 30 ] D. R. Soll, “Computer-assisted three-dimensional reconstruction and motion analysis of living, crawling cells,” Computerized Medical Imaging and Graphics, vol.23, no.1, pp.3-14, Jan. 1999.
[ 31 ] N.Otsu , “A Threshold Selection Method from Gray-level Histogram,” IEEE transactions on System , Man , and Cybernetics , vol.9, no.1, pp.62-66, Jan. 1979
[ 32 ] Z. Liang , R.J. Jaszczak and R.E. Coleman , “Parameter Estimation of Finite Mixtures Using the EM Algorithm and Information Criteria with Application to Medical Image Processing,” IEEE Transactions on Nuclear Science , vol.39 , no.4 , pp.1126-1133, 1992
[ 33 ] S. E. Chen and R. E. Parent, “Shape averaging and its applications to industrial design,” IEEE Computer Graphics and Applications, vol. 9, no.1, pp.47-54, Jan. 1989.