基于小波变换的乳腺微钙化辅助诊断算法研究
|
摘要
乳腺癌是女性最常见的癌症之一,且它在全球的发病率和死亡率正逐年增加。早期发现、早期诊断和早期治疗是降低乳腺癌死亡率的关键。乳腺X线影像是一种早期检测乳腺癌的主要手段。微钙化作为乳腺癌的早期表征,由于它极其微小、不规则,且形状和分布各异,它在乳腺X线影像中的信息只有很少部分能为人眼识别,因此微钙化的检测和诊断在临床诊断中是一件很困难的事情。随着计算机技术的飞速发展,基于乳腺X线影像的计算机辅助检测和诊断成为乳腺癌早期诊断的研究热点。目前,微钙化的计算机辅助检测已经实现了一些比较成功的商业应用,但是微钙化的计算机辅助诊断方面的研究较少,而且目前取得的良恶性分类结果远不够理想。
小波作为多尺度分析和时频分析的优良工具,它被广泛地应用到微钙化的良恶性分类中。但是前人提出的基于小波的微钙化良恶性分类方法中小波基的选取在很大程度上依赖于经验的,还没有形成一个具体和统一的标准,而且现有的微钙化良恶性分类的文献中对小波基的选取对分类结果的影响很少提及。此外,目前所用的基于小波的特征比较单一,不能取得好的分类效果。
本文主要针对微钙化的诊断分类技术进行了系统深入的研究,实现了对良性和恶性微钙化的有效分类。首先研究了小波基的特性及其对良恶性分类算法的影响;然后提出了采用单小波、多小波、方向小波和双树复数小波来进行小波变换,并提出了两组有效的多尺度纹理统计特征;最后提出了结合遗传算法(GA)和K邻域(KNN)分类器的微钙化诊断算法。
运用该算法对Nijmegen数据库中的乳腺图像进行分类,采用ROC曲线和Leave-one-out的误差估计方法对实验结果进行评价,得到了ROC曲线面积Az高达0.9469的良恶性微钙化识别率。实验结果表明对于同一个乳腺X影像数据库,本文提出的基于小波的微钙化良恶性分类方法取得的效果要明显优于前人提出的方法。通过对实验结果分析,我们还对微钙化良恶性分类方法中的小波基的选取给出一些合理的建议。
Breast cancer is the most frequently diagnosed cancer in women. Early detection and diagnosis represent a very important factor in breast cancer treatment and consequently the survival rate. Digital mammogram is considered to be the most reliable method of early detection of breast cancer. As its visual clues are subtle and varied in appearance, microcalcification detection and diagnosis is a challenging work for specialists. The computer aided diagnosis systems have been developed to aid radiologists in microcalcification detection and diagnosis. Currently, the performances reported in the literature are better for microcalcification detection than diagnosis. And the diagnosis results can’t meet the clinical needs.
Wavelet transform have been proved to be effective in classification of benign and malignant microcalcification. Little attention is paid to the selection of wavelet basis and its effect on feature extraction in current applications based on wavelet in microcalcification diagnosis. Moreover, the common features based on wavelet are too simple to get a satisfied classification results.
In this paper, we make a research on characteristics of wavelet basis and its effect on feature extraction. And we adopt the scalar wavelets, multi-wavelets, directional-wavelets and dual-tree complex wavelets to extract the muti-level information in mammogram. Two effective feature sets are proposed for feature extraction. An aided-diagnosis algorithm based on wavelet, combining with Genetic Algorithm(GA) and k-nearest-neighbor(KNN) classifier is proposed.
Receive Operating Characteristic (ROC) curve and Leave-one-out method are used to evaluate the performance of our proposed algorithm. The experimental results shown that the proposed algorithm can produce a high classification rate. Validated by the same mammographic database-Nijmegen, our algorithm is superior to previous methods. Some reasonable suggestions are also presented through analysis on experimental results.
小波作为多尺度分析和时频分析的优良工具,它被广泛地应用到微钙化的良恶性分类中。但是前人提出的基于小波的微钙化良恶性分类方法中小波基的选取在很大程度上依赖于经验的,还没有形成一个具体和统一的标准,而且现有的微钙化良恶性分类的文献中对小波基的选取对分类结果的影响很少提及。此外,目前所用的基于小波的特征比较单一,不能取得好的分类效果。
本文主要针对微钙化的诊断分类技术进行了系统深入的研究,实现了对良性和恶性微钙化的有效分类。首先研究了小波基的特性及其对良恶性分类算法的影响;然后提出了采用单小波、多小波、方向小波和双树复数小波来进行小波变换,并提出了两组有效的多尺度纹理统计特征;最后提出了结合遗传算法(GA)和K邻域(KNN)分类器的微钙化诊断算法。
运用该算法对Nijmegen数据库中的乳腺图像进行分类,采用ROC曲线和Leave-one-out的误差估计方法对实验结果进行评价,得到了ROC曲线面积Az高达0.9469的良恶性微钙化识别率。实验结果表明对于同一个乳腺X影像数据库,本文提出的基于小波的微钙化良恶性分类方法取得的效果要明显优于前人提出的方法。通过对实验结果分析,我们还对微钙化良恶性分类方法中的小波基的选取给出一些合理的建议。
Breast cancer is the most frequently diagnosed cancer in women. Early detection and diagnosis represent a very important factor in breast cancer treatment and consequently the survival rate. Digital mammogram is considered to be the most reliable method of early detection of breast cancer. As its visual clues are subtle and varied in appearance, microcalcification detection and diagnosis is a challenging work for specialists. The computer aided diagnosis systems have been developed to aid radiologists in microcalcification detection and diagnosis. Currently, the performances reported in the literature are better for microcalcification detection than diagnosis. And the diagnosis results can’t meet the clinical needs.
Wavelet transform have been proved to be effective in classification of benign and malignant microcalcification. Little attention is paid to the selection of wavelet basis and its effect on feature extraction in current applications based on wavelet in microcalcification diagnosis. Moreover, the common features based on wavelet are too simple to get a satisfied classification results.
In this paper, we make a research on characteristics of wavelet basis and its effect on feature extraction. And we adopt the scalar wavelets, multi-wavelets, directional-wavelets and dual-tree complex wavelets to extract the muti-level information in mammogram. Two effective feature sets are proposed for feature extraction. An aided-diagnosis algorithm based on wavelet, combining with Genetic Algorithm(GA) and k-nearest-neighbor(KNN) classifier is proposed.
Receive Operating Characteristic (ROC) curve and Leave-one-out method are used to evaluate the performance of our proposed algorithm. The experimental results shown that the proposed algorithm can produce a high classification rate. Validated by the same mammographic database-Nijmegen, our algorithm is superior to previous methods. Some reasonable suggestions are also presented through analysis on experimental results.
引文
[1] American Cancer Society. Cancer Facts and Figures 2008. Atlanta: American Cancer Society, 2008
[2] http://www.cnm21.com/xinwen2/041025_004.htm
[3] Giger M L. Overview of computer-aided diagnosis in breast imaging [C]. Proceedings of the first International Workshop on Computer-Aided Diagnosis. USA: 1998, 167-176.
[4] Giger M L, Huo Z, Kupinski M. Computer-aided diagnosis in mammography, in Handbook of Medical Imaging [M]. WashingtonDC: SPIE, 2000:915-986.
[5] Huo Z, Giger M L, Vyborny C J. Computerized analysis of multiple-mammographic views: potential usefulness of special view mammograms in computer-aided diagnosis[J]. IEEE Transaction on Medical Imaging, 2001, 20(12): 1285-1292.
[6] Cheng H D, Yui M L, Freimanis R I. A novel approach to microcalcification detection using fuzzy logic technique[J]. IEEE Transaction on Medical Imaging, 1998, 17(3):442-450.
[7] Sampat M P, Markey M K, Bovik A C. Computer-Aided detection and diagnosis in Mammogram, in Handbook of Image and Video Processing [M]. America: Academic Press, 2005: 1195-1217.
[8] Sickles E A. Breast calcifications: mammographic evaluation [J]. Radiology, 1986, 160(2):289-293.
[9] Shen L, Rangayyan R M, Desautels J E L. Application of shape analysis to mammographic calcifications[J]. IEEE Transaction on Medical Imaging, 1994, 13(2):263-274.
[10] Kallergi M. Computer-aided diagnosis of mammographic microcalcification clusters [J]. Medical Physis, 2004, 31(2):314-326.
[11] Veldkamp W J H, Karssemeijer N, Otten J D M. Automated classification of clustered microcalcifications into malignant and benign types[J]. Medical Physics, 2000, 27(11):2600-2608.
[12] Dhawan A P, Chitre Y, Kaiser-Bonasso C. Analysis of mammographic microcalcifications using gray-level image structure features [J]. IEEE Transaction on Medical Imaging, 1996, 15(3):246-259.
[13] Kramer D, Aghdasi F. Classification of microcalcifications in digitized mammograms using multiscale statistical texture analysis[C]. Proceedings of the South African Symposium on Communications and Signal Processing, 1998:121–126.
[14] Hamid S Z, Farshid R R, Siamak P N D. Comparison of multiwavelet, wavelet, Haralick, and shape features for microcalcification classification in mammograms[J]. Pattern Recognition, 2004, 37(10):1973-1986.
[15] Mousa R, Munib Q, Moussa A. Breast cancer diagnosis system based on wavelet analysis andfuzzy-neural[J]. Expert Systems with Application, 2005, 28(4):713-723.
[16]吴恩惠.中华影像医学-乳腺卷[M].北京:人民卫生出版社, 2002:116-117.
[17]孟悛非.医学影像学[M].北京:高等教育出版社, 2004:3-4.
[18] Mallat S. Multiresolution approximations and wavelet orthonormal bases of L2(R)[J]. Transaction of American Mathematical Society, 1989, 315(1):69-87.
[19] Mallat S. A theory for multiresolution signal decomposition: the wavelet representation [J]. IEEE Transaction on Pattern Analysis and Machine, 1989, 11(7):674-693.
[20] Mallat S. Multifrequency channel decomposition of images and wavelet models [J]. IEEE Transaction on Signal and Processing, 1989, 37(12):2091-2110.
[21] Mallat S. Zero-crossing of a wavelet transform[J]. IEEE Transaction on Information Theory, 1991, 37(8):1019-1033.
[22] Lang M, Guo H, Odegard J E, Burrus C S. Noise reduction using an undecimated discrete wavelet transform[J]. IEEE Signal Processing Letters, 1996, 3(1):10-12.
[23] Coifman R R, Donoho D L. Translation-invariant denoising [J]. Wavelets and Statistic, 1995: 125-150.
[24] Karssemeijer N. Adaptive noise equalization and recognition of microcalcification clusters in mammograms [J].International Journal of Pattern Recognition and Artificial Intelligence, 1993, 7(6):1357-1376.
[25] Veldkamp W J H, Karssemeijer N. Normalization of Local Contrast in Mammograms [J]. IEEE Transaction on Medical Imaging, 2000, 19(7):731-738.
[26] McLoughlin K, Bones P, Karssmeijer N. Noise equalization for detection of microcalcification clusters in direct digital mammogram images [J]. IEEE Transaction on Medical Imaging, 2004, 23(3):313-320.
[27] Verma B, Zakos J. A computer-aided diagnosis system for digital mammograms based on fuzzy-neural and feature extraction techniques [J]. IEEE Transaction on Information Technology in Biomedicine, 2001, 5 (1): 46–54.
[28] Shen L, Rangayyan R M, Desautels J E L. Application of shape analysis to mammographic calcifications [J]. IEEE Transaction on Medical Imaging, 1994, 13 (2): 263–274.
[29] Dhawan A P, Chitre Y, Moskowitz M, Gruenstein E. Classification of mammographic microcalcifiation and structural features using an artificial neural network[J]. Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1991, 13(3): 1105–1106.
[30] Chan H P, Sahiner B, Lam K L, Petrick N, Helvie M A, Goodsitt M M, Adler D D.Computerized analysis of mammographic microcalcifications in morphological and texture feature spaces[J].Medical Physics, 1998, 25(10):2007-2019.
[31] Zadeh H S, Nezhad S P, Rad F R. Shape-based and texture-based feature extraction for classification of microcalcification in mammograms[J]. Proceedings of the SPIE Medical Imaging Conference, 2001, ():301–310.
[32] Jiang Y, Nishikawa R M, Wolverton D E, Metz C E, Schmidt R A, Doi K. Computerized classification of malignant and benign clustered microcalcifications in mammograms[C].Proceedings of 19th International Conference-IEEE/EMBS, 1997, 521–523.
[33] Patrocinio A C, Schiabel H, Benatti R H, Goes C E, Nunes F L S. Investigation of clustered microcalcification features for an automated classifier as part of a mammography CAD scheme[C]. Proceedings of the 22nd Annual EMBS International Conference, 2000, 1203–1205.
[34] Kramer D, Aghdasi F. Texture analysis techniques for the classification of microcalcifications in digitized mammograms [C]. Proceedings of the 1999 Fifth IEEE AFRICON Conference Electrotechnical Service for Africa, 1999, 395–400.
[35] Laine A F, Schuler S, Fan J, Huda W. Mammographic feature enhancement by multiscale analysis[J]. IEEE Transaction on Medical Imaging, 1994, 13(4):725-740.
[36] Dhawan A P, Chitre Y, Bonasso C, Wheeler K. Radial-basis-function based classification of mammographic microcalcifications using texture features[C]. IEEE Proceedings of the 1995 IEEE Engineering in Medicine and Biology 17th Annual Conference and 21st Canadian Medical and Biological Engineering Conference, 1995, 535–536.
[37] Kocur C M, Gogers S K, Myers L R, Burns T, Kabrisky M, HoImeister J W, Baver K W, Steppe J M. Using neural networks to select wavelet features for breast cancer diagnosis[J]. IEEE Engeneering in Medicine and Biolgy Magazine, 1996, 15 (3): 95–102.
[38] Dhawan A P, Chitre Y, Moskowitz M. Artificial neural network based classification of mammographic microcalcifications using image structure features [C]. Proceedings of the Annual Conference on Engineering in Medicine and Biology, 1993, 50-52.
[39] Lee C S, Kim J K, Park H W. Computer-aided diagnostic system for breast cancer by detecting microcalcification[C]. Proceeding of SPIE, 1998, 615–626.
[40] Haralick R M. Statistical and structural approaches to texture [J]. Proceedings of IEEE, 1979, 67(5): 45-69.
[41] Mallat S. A wavelet tour of signal processing[M]. San Diego: Academic Press. 1998.
[42] Lebrun J, Vetterli M. High-order balanced multiwavelests: Theory, Factorization, and Design [J]. IEEE Transaction on Signal Processing, 2001, 49(9):1918-1930.
[43] Y Lu, Do M N. A directional extension for multidimensional wavelet transforms. http://www.ifp.uiuc.edu/~minhdo/publications/DEW.pdf
[44] Kingsbury N G. Image processing with complex wavelets[J]. Philosophical Transactions: Mathematical, Physical and Enginerring Sciences, 1999, 357(1760): 2543-2560.
[45] Mini M G, Thomas T. A neural network method for mammogram analysis based on statistical features. TENCON, 2003, 4(4):1489-1492.
[46] Antonie M L, Zaiane O R, Coman A. Application of data mining techniques for medical image[C]. Proceedings of the Second International Workshop on Multimedia Data Mining, 2001, 94-101.
[47] Langley P. Selection of relevant features in machine learning[C]. Proceedings of the AAAI Fall Symposium in Relevance, 1994, 140-144.
[48] Woods K S, Doss C C, Bowyer K W, Solka J L, Priebe C E, Kegelmeyer W P. Comparative evaluation of pattern recognition techniques for detection of microcalcifications inmammography[J]. International Journal of Pattern Recognition and Artificial Intelligence, 1993, 7(6):1417–1436.
[49] Li H D, Kallergi M, Clarke L P, Jain V K, Clark R A. Markov random field for tumor detection in digital mammography[J]. IEEE Transaction on Medical Imaging, 1995, 14(3): 565–576.
[50] Zheng L, Chan A K. An artificial intelligent algorithm for tumor detection in screening mammogram[J]. IEEE Transaction on Medical Imaging, 2001, 20 (17):559–567.
[51] Papadopoulos A, Fotiadis D I, Likas A. Characterization of clustered microcalcifications in digitized mammograms using neural networks and support vector machines[J]. Artificial Intelligence in Medicine, 2005, 34 (2):141-150.
[52] Soucy P, Mineau G W. A simple KNN algorithm for text categorization[C]. Proceedings of the IEEE International Conference on Data Mining, 2001, ():647–648.
[53] Chen Y Q, Nixon M S, Damper R I. Implementing the k-nearest neighbour rule via a neural network[J]. Proceedings of the IEEE International Conference on Neural Networks, 1995, 1(1):136-140.
[54] Kuncheva L I. Fitness function in editing knn reference set by genetic algorithms [J]. Pattern Recognition, 1997, 30(6):1041-1049.
[55]陈树越,路宏年.X射线数字成象噪声特性及噪声消除方法研究[J].无损检测, 2001,23(1): 9-12.
[56] Rashed E A, Ismail I A, Zaki S I. Multiresolution mammogram analysis in multilevel decomposition [J]. Pattern Recognition Letters, 2007, 28(2):286-292.
[57]李春明.乳腺X线影像中钙化点检测方法的研究(D),重庆:重庆大学,2004.
[58] Ferreira C B R, Borges D L. Automated mammogram classification using a multiresolution pattern recognition approach [C].XIV Brazilian Symposium on Computer Graphics and Image Processing (SIBGRAPI'01), 2001, 76-80.
[59] Donovan G, Geronimo J, Hardin D, Massopust P. Construction of orthogonal wavelets using fractal interpolation function[J]. Soc. Ind. Application Math. Janual on Mathematical Analysis, 1996, 27:1158-1192.
[60] Chui C K, Lian J. A study on orthonormal multi-wavelet [J]. Applied Numerical. Mathematics, 1996, 20(3):273-298.
[61] Shen L, Tan H H, Tham J Y. Symmetric-antisymmetric orthonormal multiwavelets and related scalar wavelets[J].Applied and Computational Harmonic. Analysis, 2000, 8(3):258-279.
[62]何胜名.多小波在微钙化良恶性分类中的应用研究(D),湘潭大学,2005.
[2] http://www.cnm21.com/xinwen2/041025_004.htm
[3] Giger M L. Overview of computer-aided diagnosis in breast imaging [C]. Proceedings of the first International Workshop on Computer-Aided Diagnosis. USA: 1998, 167-176.
[4] Giger M L, Huo Z, Kupinski M. Computer-aided diagnosis in mammography, in Handbook of Medical Imaging [M]. WashingtonDC: SPIE, 2000:915-986.
[5] Huo Z, Giger M L, Vyborny C J. Computerized analysis of multiple-mammographic views: potential usefulness of special view mammograms in computer-aided diagnosis[J]. IEEE Transaction on Medical Imaging, 2001, 20(12): 1285-1292.
[6] Cheng H D, Yui M L, Freimanis R I. A novel approach to microcalcification detection using fuzzy logic technique[J]. IEEE Transaction on Medical Imaging, 1998, 17(3):442-450.
[7] Sampat M P, Markey M K, Bovik A C. Computer-Aided detection and diagnosis in Mammogram, in Handbook of Image and Video Processing [M]. America: Academic Press, 2005: 1195-1217.
[8] Sickles E A. Breast calcifications: mammographic evaluation [J]. Radiology, 1986, 160(2):289-293.
[9] Shen L, Rangayyan R M, Desautels J E L. Application of shape analysis to mammographic calcifications[J]. IEEE Transaction on Medical Imaging, 1994, 13(2):263-274.
[10] Kallergi M. Computer-aided diagnosis of mammographic microcalcification clusters [J]. Medical Physis, 2004, 31(2):314-326.
[11] Veldkamp W J H, Karssemeijer N, Otten J D M. Automated classification of clustered microcalcifications into malignant and benign types[J]. Medical Physics, 2000, 27(11):2600-2608.
[12] Dhawan A P, Chitre Y, Kaiser-Bonasso C. Analysis of mammographic microcalcifications using gray-level image structure features [J]. IEEE Transaction on Medical Imaging, 1996, 15(3):246-259.
[13] Kramer D, Aghdasi F. Classification of microcalcifications in digitized mammograms using multiscale statistical texture analysis[C]. Proceedings of the South African Symposium on Communications and Signal Processing, 1998:121–126.
[14] Hamid S Z, Farshid R R, Siamak P N D. Comparison of multiwavelet, wavelet, Haralick, and shape features for microcalcification classification in mammograms[J]. Pattern Recognition, 2004, 37(10):1973-1986.
[15] Mousa R, Munib Q, Moussa A. Breast cancer diagnosis system based on wavelet analysis andfuzzy-neural[J]. Expert Systems with Application, 2005, 28(4):713-723.
[16]吴恩惠.中华影像医学-乳腺卷[M].北京:人民卫生出版社, 2002:116-117.
[17]孟悛非.医学影像学[M].北京:高等教育出版社, 2004:3-4.
[18] Mallat S. Multiresolution approximations and wavelet orthonormal bases of L2(R)[J]. Transaction of American Mathematical Society, 1989, 315(1):69-87.
[19] Mallat S. A theory for multiresolution signal decomposition: the wavelet representation [J]. IEEE Transaction on Pattern Analysis and Machine, 1989, 11(7):674-693.
[20] Mallat S. Multifrequency channel decomposition of images and wavelet models [J]. IEEE Transaction on Signal and Processing, 1989, 37(12):2091-2110.
[21] Mallat S. Zero-crossing of a wavelet transform[J]. IEEE Transaction on Information Theory, 1991, 37(8):1019-1033.
[22] Lang M, Guo H, Odegard J E, Burrus C S. Noise reduction using an undecimated discrete wavelet transform[J]. IEEE Signal Processing Letters, 1996, 3(1):10-12.
[23] Coifman R R, Donoho D L. Translation-invariant denoising [J]. Wavelets and Statistic, 1995: 125-150.
[24] Karssemeijer N. Adaptive noise equalization and recognition of microcalcification clusters in mammograms [J].International Journal of Pattern Recognition and Artificial Intelligence, 1993, 7(6):1357-1376.
[25] Veldkamp W J H, Karssemeijer N. Normalization of Local Contrast in Mammograms [J]. IEEE Transaction on Medical Imaging, 2000, 19(7):731-738.
[26] McLoughlin K, Bones P, Karssmeijer N. Noise equalization for detection of microcalcification clusters in direct digital mammogram images [J]. IEEE Transaction on Medical Imaging, 2004, 23(3):313-320.
[27] Verma B, Zakos J. A computer-aided diagnosis system for digital mammograms based on fuzzy-neural and feature extraction techniques [J]. IEEE Transaction on Information Technology in Biomedicine, 2001, 5 (1): 46–54.
[28] Shen L, Rangayyan R M, Desautels J E L. Application of shape analysis to mammographic calcifications [J]. IEEE Transaction on Medical Imaging, 1994, 13 (2): 263–274.
[29] Dhawan A P, Chitre Y, Moskowitz M, Gruenstein E. Classification of mammographic microcalcifiation and structural features using an artificial neural network[J]. Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 1991, 13(3): 1105–1106.
[30] Chan H P, Sahiner B, Lam K L, Petrick N, Helvie M A, Goodsitt M M, Adler D D.Computerized analysis of mammographic microcalcifications in morphological and texture feature spaces[J].Medical Physics, 1998, 25(10):2007-2019.
[31] Zadeh H S, Nezhad S P, Rad F R. Shape-based and texture-based feature extraction for classification of microcalcification in mammograms[J]. Proceedings of the SPIE Medical Imaging Conference, 2001, ():301–310.
[32] Jiang Y, Nishikawa R M, Wolverton D E, Metz C E, Schmidt R A, Doi K. Computerized classification of malignant and benign clustered microcalcifications in mammograms[C].Proceedings of 19th International Conference-IEEE/EMBS, 1997, 521–523.
[33] Patrocinio A C, Schiabel H, Benatti R H, Goes C E, Nunes F L S. Investigation of clustered microcalcification features for an automated classifier as part of a mammography CAD scheme[C]. Proceedings of the 22nd Annual EMBS International Conference, 2000, 1203–1205.
[34] Kramer D, Aghdasi F. Texture analysis techniques for the classification of microcalcifications in digitized mammograms [C]. Proceedings of the 1999 Fifth IEEE AFRICON Conference Electrotechnical Service for Africa, 1999, 395–400.
[35] Laine A F, Schuler S, Fan J, Huda W. Mammographic feature enhancement by multiscale analysis[J]. IEEE Transaction on Medical Imaging, 1994, 13(4):725-740.
[36] Dhawan A P, Chitre Y, Bonasso C, Wheeler K. Radial-basis-function based classification of mammographic microcalcifications using texture features[C]. IEEE Proceedings of the 1995 IEEE Engineering in Medicine and Biology 17th Annual Conference and 21st Canadian Medical and Biological Engineering Conference, 1995, 535–536.
[37] Kocur C M, Gogers S K, Myers L R, Burns T, Kabrisky M, HoImeister J W, Baver K W, Steppe J M. Using neural networks to select wavelet features for breast cancer diagnosis[J]. IEEE Engeneering in Medicine and Biolgy Magazine, 1996, 15 (3): 95–102.
[38] Dhawan A P, Chitre Y, Moskowitz M. Artificial neural network based classification of mammographic microcalcifications using image structure features [C]. Proceedings of the Annual Conference on Engineering in Medicine and Biology, 1993, 50-52.
[39] Lee C S, Kim J K, Park H W. Computer-aided diagnostic system for breast cancer by detecting microcalcification[C]. Proceeding of SPIE, 1998, 615–626.
[40] Haralick R M. Statistical and structural approaches to texture [J]. Proceedings of IEEE, 1979, 67(5): 45-69.
[41] Mallat S. A wavelet tour of signal processing[M]. San Diego: Academic Press. 1998.
[42] Lebrun J, Vetterli M. High-order balanced multiwavelests: Theory, Factorization, and Design [J]. IEEE Transaction on Signal Processing, 2001, 49(9):1918-1930.
[43] Y Lu, Do M N. A directional extension for multidimensional wavelet transforms. http://www.ifp.uiuc.edu/~minhdo/publications/DEW.pdf
[44] Kingsbury N G. Image processing with complex wavelets[J]. Philosophical Transactions: Mathematical, Physical and Enginerring Sciences, 1999, 357(1760): 2543-2560.
[45] Mini M G, Thomas T. A neural network method for mammogram analysis based on statistical features. TENCON, 2003, 4(4):1489-1492.
[46] Antonie M L, Zaiane O R, Coman A. Application of data mining techniques for medical image[C]. Proceedings of the Second International Workshop on Multimedia Data Mining, 2001, 94-101.
[47] Langley P. Selection of relevant features in machine learning[C]. Proceedings of the AAAI Fall Symposium in Relevance, 1994, 140-144.
[48] Woods K S, Doss C C, Bowyer K W, Solka J L, Priebe C E, Kegelmeyer W P. Comparative evaluation of pattern recognition techniques for detection of microcalcifications inmammography[J]. International Journal of Pattern Recognition and Artificial Intelligence, 1993, 7(6):1417–1436.
[49] Li H D, Kallergi M, Clarke L P, Jain V K, Clark R A. Markov random field for tumor detection in digital mammography[J]. IEEE Transaction on Medical Imaging, 1995, 14(3): 565–576.
[50] Zheng L, Chan A K. An artificial intelligent algorithm for tumor detection in screening mammogram[J]. IEEE Transaction on Medical Imaging, 2001, 20 (17):559–567.
[51] Papadopoulos A, Fotiadis D I, Likas A. Characterization of clustered microcalcifications in digitized mammograms using neural networks and support vector machines[J]. Artificial Intelligence in Medicine, 2005, 34 (2):141-150.
[52] Soucy P, Mineau G W. A simple KNN algorithm for text categorization[C]. Proceedings of the IEEE International Conference on Data Mining, 2001, ():647–648.
[53] Chen Y Q, Nixon M S, Damper R I. Implementing the k-nearest neighbour rule via a neural network[J]. Proceedings of the IEEE International Conference on Neural Networks, 1995, 1(1):136-140.
[54] Kuncheva L I. Fitness function in editing knn reference set by genetic algorithms [J]. Pattern Recognition, 1997, 30(6):1041-1049.
[55]陈树越,路宏年.X射线数字成象噪声特性及噪声消除方法研究[J].无损检测, 2001,23(1): 9-12.
[56] Rashed E A, Ismail I A, Zaki S I. Multiresolution mammogram analysis in multilevel decomposition [J]. Pattern Recognition Letters, 2007, 28(2):286-292.
[57]李春明.乳腺X线影像中钙化点检测方法的研究(D),重庆:重庆大学,2004.
[58] Ferreira C B R, Borges D L. Automated mammogram classification using a multiresolution pattern recognition approach [C].XIV Brazilian Symposium on Computer Graphics and Image Processing (SIBGRAPI'01), 2001, 76-80.
[59] Donovan G, Geronimo J, Hardin D, Massopust P. Construction of orthogonal wavelets using fractal interpolation function[J]. Soc. Ind. Application Math. Janual on Mathematical Analysis, 1996, 27:1158-1192.
[60] Chui C K, Lian J. A study on orthonormal multi-wavelet [J]. Applied Numerical. Mathematics, 1996, 20(3):273-298.
[61] Shen L, Tan H H, Tham J Y. Symmetric-antisymmetric orthonormal multiwavelets and related scalar wavelets[J].Applied and Computational Harmonic. Analysis, 2000, 8(3):258-279.
[62]何胜名.多小波在微钙化良恶性分类中的应用研究(D),湘潭大学,2005.