臺灣博碩士論文加值系統

利用電腦視覺系統來輔助分析醫學影像將有助於醫生做出更精確的診斷。而對於電腦視覺系統，影像的特徵擷取是相當重要的一步。本論文的主要目的為探討超音波肝臟影像的特徵擷取。因此，我們提出以M-頻段小波轉換為基礎的碎形特徵向量來對正常肝、肝硬化、肝癌等三類的超音波肝臟影像辨識。而所提出的特徵擷取法則是架構在空間-頻率的分解及碎形幾何上。同時，我們也提出以不同的紋理量測與濾波器組為基礎的分類法則並且測試之。在三類超音波肝臟影像的辨識實驗中，以M-頻段小波轉換為基礎的碎形特徵向量對於辨識的結果是非常令人鼓舞的。由於在超音波肝臟影像中，最明顯的特徵為粗糙度（roughness），其中肝癌影像較肝硬化影像來的粗糙。因此，我們提出一階層式的分類器，首先辨識正常肝臟與不正常肝臟，其辨識率為96.7％；接著辨識肝硬化與肝癌，其辨識率為93.6％。此外，我們也探討如何選取一合適的特徵向量並做一些比較性的實驗。
在碎形幾何中，碎形維度（fractal dimension）是相當重要的一個特徵。因此，在監督式的辨識實驗中，我們對碎形維度的計算，提出一更為有效和穩健的計算方法。我們所提出的計算方法是以「積木計數」為基礎並加以改良。由於原始方法「積木計數」是相當容易受雜訊干擾的，以致產生許多明顯的平坦區域（plateau）而使得碎形維度的計算值被低估。因此，從實驗的結果，可證實我們所提出的碎形維度的計算方法是非常穩健且相當有效率的。
最後，我們應用所提出的多重解析度碎形特徵向量來切割出超音波肝臟影像中可能為病變的區域。不同的病變肝臟影像的切割實驗顯示出，以多重解析度為基礎的碎形特徵向量可以有效的描述肝臟組織影像的紋理。因此，我們可以利用所提出的非監督式的辨識切割法則來建立一套可定量描述的自動化電腦輔助診斷分析系統。同時，為了提升年輕的臨床醫生對超音波肝臟影像的視覺辨識能力，我們可利用所提出的非監督式的辨識切割法則來建立一套離線（off-line）的學習系統以探討超音波肝臟影像的視覺辨識準則。

The focus of this dissertation is the feature extraction of ultrasonic liver images. The feature extraction is essential in a computer vision system for diagnosis of medical images. We propose a fractal feature vector based on M-band wavelet transform to classify ultrasonic liver images─normal liver, cirrhosis, and hepatoma. The proposed feature extraction algorithm is based on the spatial-frequency decomposition and fractal geometry. And, various classification algorithms based on respective texture measurements and filter banks are presented and tested. Classifications for the three sets of ultrasonic liver images reveal that the fractal feature vector based on M-band wavelet transform is trustworthy. A hierarchical classifier, which is based on the proposed feature extraction algorithm is at least 96.7% accurate in the distinction between normal and abnormal liver images and is at least 93.6% accurate in the distinction between cirrhosis and hepatoma liver images. Additionally, the criterion for feature selection is specified and employed for performance comparisons herein.
In supervised classification, we also propose a modified computation of fractal dimension since the estimation of the fractal dimension is crucial in fractal geometry. The adopted estimation approach is based on box-counting. However, the scheme, which is easily disturbed by noise, produced many non-negligible plateaus that cause underestimate. A more robust and efficient computation of the fractal dimension is verified from experimental results.
Finally, we applied the proposed multiresolution fractal feature vector to segment suspicious abnormal regions of ultrasonic liver images. Segmentation of various liver diseases reveals that the fractal feature vector based on multiresolution analysis is reliable. A quantitative characterization based on the proposed unsupervised segmentation algorithm can be utilized to establish an automatic computer-aided diagnostic system. As well, to increase the visual interpretation capability of ultrasonic liver image for junior physicians, an off-line learning system can be developed to investigate the visual criteria.

Chapter 1 Introduction
1.1 Motivation
1.2 Texture Analysis
1.3 Fractal Geometry
1.4 Multiresolution Analysis Based on M-band Wavelet Transform
1.5 Contribution of this Dissertation
1.6 Organization of this Dissertation
Chapter 2 Feature Extraction Approach
2.1 Introduction
2.2 Robust calculation of fractal dimension
2.3 General Multiresolution Analysis based on M-band Wavelet Transform
2.4 The Multiresolution Fractal Feature Vector
2.5 Pattern Classification Techniques
2.5.1 Statistical Classifiers
2.5.2 Neural Network Classifiers
2.6 Experimental Results
2.6.1 Experimental data set of natural textures
2.6.2 Classification results of natural textures
2.6.3 The performance of proposed computation of fractal dimension
2.6.4 Sensitivity to noisy data
2.6.5 Comparison with energy feature vector
2.6.6 Comparison with wavelet frame
2.6.7 Comparison under exchange of training set and test set
2.6.8 Ultrasonic liver images classification
2.7 Summary
Chapter 3 Supervised Classification of Ultrasonic Liver Images
3.1 Introduction
3.2 The Fractal Feature Vector based on M-band Wavelet Transform
3.3 Other Measurements of Texture Feature
3.4 The Feature Vector based on Various Filter Bank
3.5 A Hierarchical Classifier
3.6 Experimental Results
3.6.1 Image Acquisition
3.6.2 Using Orthogonal Bases at the First Level of Wavelet
Analysis System
3.6.3 Using Biorthogonal Bases at the First Level of Wavelet Analysis System
3.6.4 Comparison with Gabor Filter Bank
3.6.5 Discussion
3.7 Summary
Chapter 4 Unsupervised Segmentation of Ultrasonic Liver Images
4.1 Introduction
4.2 .The Multiresolution Fractal Feature Vector based on M-band Wavelet Transform
4.3 Unsupervised segmentation algorithm
4.4 Segmentation Results
4.4.1 Segmentation of natural texture image
4.4.2 Segmentation of ultrasonic liver image
4.4.3 Discussion
4.5 Summary
Chapter 5 Summary and Conclusions
5.1 Summary and Conclusions
5.2 Suggestions for Future Research

[1] M. F. Insana, R. F. Wagner, B. S. Garra, D. G. Brown, and T. H. Shawker, “Analysis of Ultrasound image texture via generalized Rician statistics,” Opt. Eng., vol. 25, pp. 743-748, 1986.
[2] U. Reath, D. Schlaps, and B. Limberg, “Diagnostic accuracy of computerized B-scan texture analysis and conventional ultrasonography in diffuse parenchymal and malignant liver disease,” J. Clin. Ultrasound, vol. 13, pp. 87-99, 1985.
[3] N. M. Botros, “ A microprocessor-based pattern recognition algorithm for in-vivo tissue differentiation,” J. Clin. Eng., vol. 13, pp.115-120, 1988.
[4] R. Momenan, M. H. Loew, M. F. Insana, R. F. Wagner, and B. S. Garra, “Application of pattern recognition techniques in ultrasound tissue characterization,” 10th Int. Conf. Pattern Recognition, vol. 1, pp. 608-612, 1990.
[5] M. F. Insana, R. F. Wagner, B. S. Garra, R. Momenan, and T. H. Shawker, “Pattern recognition methods for optimizing multivariate tissue signatures in diagnostic ultrasound,” Ultrasound. Imaging, vol. 8, pp165-180, 1986.
[6] B. S. Garra, M. F. Insana, T. H. Shawker, R. F. Wagner, M. Bradford and M. Russell, “Quantitative ultrasonic detection and classification of diffuse liver disease comparison with human observer performance,” Invest. Radiology, vol. 24, pp.196-203, 1989.
[7] R. F. Wagner, M. F. Insana, and G. Brown, “Unified approach to the detection and classification of speckle texture in diagnostic ultrasound,” Opt. Eng., vol. 25, pp. 743-748, 1986.
[8] R. Momenan, M. F. Insana, R. F. Wagner, B. S. Garra, and M. H. Loew, “Application of clutter analysis and unsupervised learning to multivariate tissue characterization,” J. Clin. Eng., vol. 13, pp.455-461, 1988.
[9] R. Momenan, R. F. Wagner, B. S. Garra, M. H. Loew, and M. F. Insana, “Image staining and differential diagnosis of ultrasound scans based on the Mahalanobis distance,” IEEE Trans. Medical Imaging, vol. 11, pp. 37-47, June 1994.
[10] K. Ogawa, M. Fukushima, K. Kubota, and N. Hisa, “Computer-aided Diagnostic System for Diffuse Liver Diseases with Ultrasonography by Neural Network,” IEEE Trans. Nuclear Science, vol. 45, no. 6, pp. 3069-3074, Dec. 1998.
[11] G. J. W Simon, E. E. Jane, B. Nigal, E. H. Margaret, E. B. Joe, and W. David, “An ultrasound scoring system for the diagnosis of liver disease in cystic fibrosis,” Journal of Hepatology, vol. 22, pp. 513-521, 1995.
[12] G. F. Vawter and H. Shwachman, “Cystic fibrosis in adults: an autopsy study,” Pathol. Annu., vol. 14, pp.357-382, 1979.
[13] N. I. Sandford, P. Walsh, C. Matis, H. Baddeley and L. W. Powell, “Is ultrasonography useful in the development of diffuse parenchyma liver disease,” Gastroenterology, vol. 9, pp. 186-191, 1985.
[14] C. M. Wu, Y. C. Chen and K. S. Hsieh, “Texture features for classification of ultrasonic liver images,” IEEE Trans. Medical Imaging, vol. 11, pp. 141-152, June 1992.
[15] C. M. Wu and Y. C. Chen, “Multi-threshold dimension vector for texture analysis and its application to liver tissue classification,” Pattern Recognition, vol. 26, no. 1, pp. 137-144, Jan. 1993
[16] A. Mojsilović, M. Popović, S. Marković, and M. Krstić, “Characterization of Visually Similar Diffuse Diseases from B-Scan Liver Images Using Nonseparable Wavelet Transform,” IEEE Trans. Medical Imaging, vol. 17, no. 4, pp. 541-549, Aug. 1998.
[17] Y. C. Chen and W. L. Lee, “Texture Classification Using Multiresolution Fractal Feature Vector,” Proc. 4th Asian Conf. On Computer Vision, pp. 204-209, 2000.
[18] W. L. Lee, Y. C. Chen and K. S. Hsieh, “Ultrasonic liver tissues classification by fractal feature vector based on M-band wavelet transform,” The 2001 IEEE International Symposium on Circuits and Systems, Vol. 2, pp. 1-4, 2001.
[19] W. L. Lee, Y. C. Chen and K. S. Hsieh, “Robust calculation of fractal dimension of images and its applications to classiflication of ultrasonic liver images and texture images,” The 2002 IEEE International Symposium on Circuits and Systems, Vol. 2, pp. 656-659, 2002.
[20] H. Sujana, S. Swarnamani, and S. Suresh, “ Application of artificial neural networks for the classification of liver lesions by image texture parameters,” Ultrasound in Med. & Biol., vol. 22, no. 9, pp. 1177-1181, 1996.
[21] S. Pavlopoulos, E. Kyriacou, D. Koutsouris, K. Blekas, a. Stafylopatis, and P. Zoumpoulis, “Fuzzy neural network-based texture analysis of ultrasonic images,” IEEE Engineering in Medicine and Biology, vol. 19, no. 1, pp. 39-47, Jan.-Feb., 2000.
[22] Y. M. Kadah, A. A. Farag, J. M. Zurada, A. M. Badawi, and A. M. Youssef, “Classification Algorithms for Quantitative Tissue Characterization of Diffuse Liver Disease from Ultrasound Images,” IEEE Trans. Medical Imaging, vol. 15, no. 4, pp. 466-478, Aug. 1996.
[23] B. J. Oosterveld, J. M. Thijssen, P. C. Hartman, and G. J. E. Rosenbusch, “Detection of diffuse liver disease by quantative echography: dependence on a priori choice of parameters,” Ultrasound in Med. & Biol. vol.19, no.1 pp.21-25, 1993.
[24] R. M. Haralick, K. Shanmugam, and I. Dinstein, “Texture features for image classification,” IEEE Trans. Systems, Man and Cybernetics, vol. 3, no. 6, pp.610-621, 1973.
[25] C. M. Wu and Y. C. Chen, “Statistical fature matrix for texture analysis,” CVGIP, vol. 54, No. 5, September, pp407-419, 1992.
[26] P. C. Chen and T. Pavlidis, “Segmentation by texture using correlation,” IEEE Trans. Pattern Anal. and Machine Intell., vol. 5, pp. 64-69, Jan. 1983.
[27] R. Chellapa, “Two-dimensional discrete Gaussian Marks random field models for image processing,” Pattern Recognition, vol. 2, pp. 79-112, 1985.
[28] K. I. Laws, Texture image segmentation, Ph.D. dissertation, Image Processing Inst., Univ. of Southern California, 1980.
[29] Campbell, F.W. and Robson, J.G., “Application of Fourier Analysis to the Visibility of Gratings,” J. Physilogy, vol.197, pp.551-566, 1968.
[30] De Valois, R.L. Albrecht D.G., and Thorell, L.G., “Spatial-Frequency Selectivity of Cells in Macaque Visual Cortex,” Vision Research, vol.22, pp.545-559,1982.
[31] B. B. Mandelbrot, Fractal Geometry of Nature. Freeman Press, San Francisco, 1982.
[32] Pentland, A.P., “Fractal based description of natural scences,” IEEE Trans. Pattern Anal. and Machine Intell., vol. PAMI-6, pp.661-674, 1984.
[33] Pentland, A.P., “Shading into texture,” Artificial Intelligence, vol. 29, pp.147-170, 1986.
[34] C. C. Chen, J. S. Daponte and M. D. Fox, “Fractal feature analysis and classification in medical imaging,” IEEE Trans. Medical Imaging, vol. 6, pp. 133-142, June 1989.
[35] S. C. Liu and S. Chang, “Dimension estimation of discrete-time fractional Brownian motion with applications to image texture classification,” IEEE Trans. Image Processing, vol. 6, no. 8, Aug. 1997.
[36] A. Rosenfeld, Multiresolution Techniques in Computer Vision, New York: Springer-Verlag, 1984.
[37] P.P. Vaidyanathan, “Multirate digital filters, filter banks, polyphase network, and applications,” in Pro. IEEE, vol. 78, Jan. 1990, pp.56-93.
[38] M. Vettterli and C. Herley, “Wavelet and filter banks: Theory and design,” IEEE Trans, Signal Processing, vol. 40, pp.2207-2232, Sept. 1992.
[39] M. Frisch and H. Messer, “Detection of a transient signal of unknown scaling and arrival time using the discrete wavelet transform,” in Proc. Int. Conf. Acoust., Speech, Signal Processing, vol. 42, pp.595-603, March 1994.
[40] S. Mallat, “Zero-crossings of a wavelet transform,” IEEE Trans. Inform. Theory, vol. 37, pp. 1019-1033, July 1991.
[41] S. Kadambe and G. F. Boudraux-bartels, “Application of the wavelet transform for pitch detection of speech signals,” IEEE Trans. Inform. Theory, vol. 38, pp.917-924, Mar. 1992.
[42] D. M. Healy and J. B. Weaver, “Two applications of wavelet transforms in magnetic resonance imaging,” IEEE Trans. Inform. Theory, vol. 38, pp.840-860, Mar. 1992.
[43] S. Mallat, “A theory for multiresolution signal decomposition: the wavelet representation,” IEEE Trans. Pattern Anal. and Machine Intell., vol. 11, pp. 674-693, July 1989.
[44] T. Chang and C. Kuo, “Texture analysis and classification with tree-structured wavelet transform,” IEEE Trans. Image Processing, vol. 2, pp.429-441, Oct. 1993.
[45] A. Laine and J. Fan, “Texture classification by wavelet packet signatures,” IEEE Trans. Pattern Anal. and Machine Intell., vol. 15, pp. 1186-1191, Nov. 1993.
[46] M. Unser, “Texture classification and segmentation using wavelet frames,” IEEE Trans. Image Processing, vol. 4, no. 11, pp. 1549-1560, Nov. 1995.
[47] D. Hubel and T.Wiesel, “Receptive fields, binocular interaction and functional architecture in the cat’s visual cortex”, J. Physilogy, vol.160, pp.551-566, 1962.
[48] D. A. Pollen and S. F. Ronner, “Visual cortical neurons as localized spatial frequency filter.” IEEE Trans. Sys., Man, Cyber., vol.13, pp.551-566, 1983.
[49] J.M. Keller, S. Chen, and R.M. Crowniver, “Texture description and segmentation through fractal geometry,” CVGIP, 45, pp.150-166, 1989.
[50] S. Chen, J.M. Keller, and R.M. Crowniver, “On the calculation of fractal features from images,” IEEE Trans. Pattern Anal. and Machine Intell., vol. 15, no. 10, pp. 1087-1090, Oct. 1993.
[51] N. Sarkar and B. B. Chaudhuri, “An efficient differential box-counting approach to compute fractal dimension of image,” IEEE Trans. Systems Man and Cybernetics, vol. 24, pp. 115-120, Jan. 1994.
[52] X. C. Jin, S. H. Ong, and Jayasooriah, “A pracital method for estimating fractal dimension,” Pattern Recognition Letter 16, 411-418 (1995).
[53] J. Feng, W-C Lin, and C-T Chen, “Fractional box counting approach to fractal dimension estimation,” Proceedings, 13th ICPR, vol. II, pp.854-858, 1996.
[54] S. Buczkowski, S. Kyriacos, F. Nekka, and L. Cartilier, “The Modified Box-Counting Method: Analysis of Some Characteristic parameters,” Pattern Recognition, vol. 31, No. 4, pp. 411-418, 1998.
[55] M. K. Biswas, T. Ghose, S. Guha, and P. K. Biswas, “Fractal dimension estimation for texture images: A parallel approach,” Pattern Recognition Letter 19, 309-313 (1998).
[56] P. Asvestas, G. K. Matsopoulos, K. S. Nikita, “Estimation of fractal dimension of images using a fixed mass approach,” Pattern Recognition Letter 20, 347-354 (1999).
[57] A. K. Bisoi and J. Mishra, “On calculation of fractal dimension of images,” Pattern Recognition Letter 22, 631-637 (2001).
[58] T. Lundahl, W. J. Ohely, S. M. Kay, and R. Siffert, “Fractional Brownian Motion: A Maximum Likelihood Estimator and Its Application to Image Texture,” IEEE Trans. Medical Imaging, vol. 5, no, 3, pp. 152-161, Sep. 1986.
[59] J. T. M. Verhoeven and J. M. Thijssen, “Potential of fractal analysis for lesion detection in echographic images,” Ultrasonic Imaging, vol.15, pp.304-323, 1993.
[60] G. Strang and T. Nguyen, Wavelets and Filter Banks. MA: Wellesley-Cambridge, 1996.
[61] C. S. Burrus, R. A. Goponath, and H. Guo, Introduction to wavelets and wavelet transform: a primer. New Jersey: Prentice-Hall, 1998.
[62] R. A. Goponath, E. Odegard, and C. S. Burrus, “Optimal wavelet representation of signals and the wavelet sampling theorem,” IEEE Trans. Circuits and systems. II, vol. 41, no. 4, pp. 262-277, April 1994.
[63] M. K. Tsatsanis and G. B. Giannakis, “ Principal component filter banks for optimal multiresolution analysis,” IEEE Trans. Signal Processing, vol. 43, no.8, pp.1766-1777, Aug. 1990.
[64] P. Steffen, P. N. Heller, R. A. Goponath, and C. S. Burrus, “Theory of Regular M-Band Wavelet Bases,” IEEE Trans. Signal Processing, vol. 41, No. 12, pp. 3497-3511, Dec. 1993.
[65] O. Rioul and M. Vetterli, “Wavelets and signal processing,” IEEE Signal Processing Mag., vol. 8, no. 4, pp. 11-38, Oct. 1991
[66] M. K. Tsatsanis and G. B. Giannakis, “Principal Component Filter Banks for Optimal Multiresolution Analysis,” IEEE Trans. Signal Processing, vol. 43, No. 8, pp. 1766-1777, Aug. 1995.
[67] S. Mallat, A Wavelet Tour of Signal Processing. Academic Press, San Diego, 1998.
[68] Y. Chitre and A. P. Dhawan, “M-band wavelet discrimination of nature textures,” Pattern Recognition, vol. 32, pp. 773-789, 1999.
[69] P. P. Vaidaynathan, Multirate Systems and Filter Banks. New Jersey: Prentice-Hall, 1993
[70] I.Daubechies, “The wavelet transform, time-frequency localization and signal analysis,” IEEE Trans. Information Theory, vol. 36, pp.961-1005, Sept. 1990.
[71] R. O. Duda and P. E. Hart, Pattern Classification and Scene Analysis. New York: Wiley, 1973.
[72] M. Nadler and E. P. Smith, Pattern Recognition Engineering. New York: Wiley, 1993
[73] K. Fukunaga, Introduction to Statistical Pattern Recognition. Academic Press, San Diego, 1990.
[74] M. T. Hagan, H. B. Demuth, and M.H. Beale, Neural Network Design, Boston, MA: PWS Publishing, 1996.
[75] D. F. Specht, “Probabilistic neural network,” Neural Networks, vol. 3, no. 1, pp. 109-118, Jan. 1990.
[76] D. F. Specht, “Probabilistic neural network and the polynomial adaline as complementary techniques for classification,” IEEE Trans. Neural Networks, vol. 1, no. 1, pp. 111-121, Mar. 1990.
[77] P. Burrascano, “Learning vector quantization for the probabilistic neural network,” IEEE Trans. Neural Networks, vol. 2, pp. 458-461, 1991.
[78] E-Liang Chen, Pau-Choo Chung, Ching-Liang Chen, Hong-Ming Tsai, and Chein-I Chang, “An automatic diagnostic system for CT liver image classification,” IEEE Trans. Biomedical Engineering, vol. 45, no. 6, June 1998.
[79] P. Brodatz, Texture: A Photographic Album for Artists and Designers. New York: Dover, 1966.
[80] A. K. Jain and F. Farrokhnia, “Unsupervised Texture Segmentation Using Gabor Filters,” Proceedings of IEEE International Conference on Systems, Man and Cybernetics, pp. 14-19, 1990.
[81] A. K. Jain and F. Farrokhnia, “Unsupervised texture segmentation using Gabor filters,” Pattern Recognition, vol. 24, no. 12, pp.1167-1186, 1991.
[82] T. Randen and J. H. Husoy, “Mutichannel filtering for image texture segmentation,” Optical Eng., vol.33, pp.2617-2625, Aug. 1994.
[83] P. J. Scheuer, “Pathologic types of hepatic tumors,” in Liver cell carcinoma, P. Bannasch, D. Keppler, and G. Weber, Eds. New York: Kluwer-Academic, p.18, 1989.
[84] J. H. Lefkowitch, “Pathologic diagnosis of liver disease,” in Hepatology: a Textbook of liver Disease, D. Zakim and T. D. Boyer, W. B. Saunders, Eds. London, England, p.719, 1990.
[85] T. Randen and J. H. Husoy, “Filtering for Texture Classification: A Comparative Study,” IEEE Trans. Pattern Anal. and Machine Intell., vol. 21, no. 4, pp. 291-310, April 1999.