摘要
2008年诺贝尔奖生理学奖的揭晓让世界的眼光又再次聚焦到了妇女第二大多发癌症——宫颈癌上。根据美国癌症协会与2007年发表的相关报告,在全球范围内,宫颈癌已经成为了女性因肿瘤死亡的第二号杀手。
同时,随着腔镜技术的不断发展,阴道镜已成为宫颈癌和癌前病变的重要辅助诊断方法之一。阴道镜其最主要的临床应用价值是,在阴道镜指示下定点活检和HPV病毒感染的检查,以及提高活检的阳性率、宫颈癌的早期诊断率和降低假阴性比例。
本文首先针对我国在阴道镜图像数据库资料保存方面的不足,提出了一种新的阴道镜图像的数据库系统的设计与实现,作用于宫颈癌早期筛查,为医疗建档、医学研究、教学、会诊、实验室资料的保存和临床的其他领域提供帮助。
其次,为了辅助提取更为有效的活检组织,本文提出了四种可选择的图像处理的方法,做到自动对于原始图像的高亮区域提取与填充、自动进行宫颈口ROI区域的分割,对醋酸白化上皮区域的特征检测与分析,标记疑似醋酸白化上皮区域以指导有效活检等。
最后,对阴道镜图像自动识别的研究作出尝试,通过对阴道镜图像的分析选取计算图像颜色信息中的特征,对这些特征进行的初步分析,应用支持向量机作为特征分类的分类器,得到其分类精度78.1%,敏感性81.1%和特异性75%的结果。其结果与原始图像分类结果相比改善明显,特征有效性获得证实。
The 2008 Nobel Prize in physiology caught the world’s eyes onto the research in uterine cervical cancer. Uterine cervical cancer has been the second most common cancer in women worldwide, according to the 2007’s report from NCI. Optical colposcopy is the primary method used to detect uterine cervical cancer or Cervical Intraepithelial Neoplasia.
Due to the insufficiency of China in colposcopy image collection , a novel image database is developed and it will be the basis of a computer-aided-diagnosis system for colposcopy. Teaching, preservation of prime data, medical research, etc. would also benefit from the setting-up of the image database.
Secondly, in order to get more effective biopsies, four alternative processing methods are proposed to segment effective region from original images, detect and analyze the features of acetowhite regions and mark candidate acetowhite regions within digital uterine cervix images automatically.
33 color features are extracted for the image analysis after processing the original images in the above methods. Support vector machine classifier is applied -- classifying accuracy is 78.1%, sensitivity is 81.1% and specificity is 75%. The result shows a considerable increase by around 15% on the accuracy which proves the validity of the feature extraction and the classifying scheme.
引文
[1] D. Parkin, et al., "Global cancer statistics, 2002," CA: a cancer journal for clinicians, vol. 55, p. 74, 2005.
[2] S. Pagliusi, "World Health Organization. Human papillomavirus infection and cervical cancer," ed.
[3]张志胜, et al., "阴道镜图谱北京:人民卫生出版社2000," ed: P.
[4] N. Munoz, et al., "Epidemiologic classification of human papillomavirus types associated with cervical cancer," New England Journal of Medicine, vol. 348, pp. 518-528, 2003.
[5] E. Burghardt, et al., Primary care colposcopy: textbook and atlas: George Thieme Verlag, 2004.
[6] F. Bosch, et al., "Prevalence of human papillomavirus in cervical cancer: a worldwide perspective," JNCI Journal of the National Cancer Institute, vol. 87, p. 796, 1995.
[7] D. Saslow, et al., "American Cancer Society Guideline for human papillomavirus (HPV) vaccine use to prevent cervical cancer and its precursors," CA: a cancer journal for clinicians, vol. 57, p. 7, 2007.
[8] J. Ferlay, et al., "GLOBOCAN 2000 [: Cancer Incidence, Mortality and Prevalence Worldwide," 2001.
[9] P. Pisani, et al., "Estimates of the worldwide mortality from eighteen major cancers in 1985. Implications for prevention and projections of future burden," International Journal of Cancer, vol. 55, pp. 891-903, 1993.
[10] A. Cinel, et al., "The accuracy of colposcopically directed biopsy in the diagnosis of cervical intraepithelial neoplasia," European journal of gynaecological oncology, vol. 11, p. 433, 1990.
[11] D. Ferris, et al., "Colposcopy quality control: Establishing colposcopy criterion standards for the National Cancer Institute ALTS trial using cervigrams," Journal of Lower Genital Tract Disease, vol. 2, p. 195, 1998.
[12] M. Coppleson, et al., "Colposcopic differentiation of abnormalities arising in the transformation zone," Obstetrics and gynecology clinics of North America, vol. 20, p. 83, 1993.
[13] M. MIKHAIL, et al., "Clinical usefulness of computerized colposcopy: image analysis and conservative management of mild dysplasia," Obstetrics & Gynecology, vol. 80, p. 5, 1992.
[14] S. Ballagh, et al., "A comparison of techniques to assess cervicovaginal irritation and evaluation of the variability between two observers," Contraception, vol. 70, pp. 241-249, 2004.
[15] M. Toglia, et al., "Evaluation of colposcopic skills in an obstetrics and gynecology residency training program," Journal of Lower Genital Tract Disease, vol. 1, p. 5, 1997.
[16] K. Nanda, et al., "Accuracy of the Papanicolaou test in screening for and follow-up of cervical cytologic abnormalities: a systematic review," Annals of Internal Medicine, vol. 132, p. 810, 2000.
[17] F. Baum, et al., "Computer-aided detection in direct digital full-field mammography: initial results," European radiology, vol. 12, pp. 3015-3017, 2002.
[18] B. Van Ginneken, et al., "Computer-aided diagnosis in chest radiography: a survey," IEEE Transactions on Medical Imaging, vol. 20, pp. 1228-1241, 2001.
[19] P. Cristoforoni, et al., "Computerized colposcopy: Results of a pilot study and analysis of its clinical relevance," OBSTETRICS AND GYNECOLOGY-NEW YORK-, vol. 85, pp. 1011-1011, 1995.
[20] B. Pogue, et al., "Image analysis for discrimination of cervical neoplasia," Journal of Biomedical Optics, vol. 5, p. 72, 2000.
[21] H. Lange, "Automatic detection of multi-level acetowhite regions in RGB color images of the uterine cervix," 2005, pp. 1004-1017.
[22] V. Van Raad, et al., "Lesion margin analysis for automated classification of cervical cancer lesions," Medical Imaging 2006: Image Processing, pp. 1647-59, 2006.
[23] S. Gordon, et al., "Content analysis of uterine cervix images: initial steps towards content based indexing and retrieval of cervigrams," 2006, pp. 1549-1556.
[24] V. Van Raad, "Active contour models-a multiscale implementation for anatomical feature delineation in cervical images," 2004, pp. 24-27.
[25] S. Yang, et al., "A multispectral digital Cervigram analyzer in the wavelet domain for early detection of cervical cancer," 2004, p. 1833.
[26] S. Gordon, et al., "Image segmentation of uterine cervix images for indexing in PACS," 2004, pp. 298–303.
[27] Q. Ji, et al., "Texture analysis for classification of cervix lesions," IEEE Transactions on Medical Imaging, vol. 19, pp. 1144-1149, 2000.
[28] C. Balas, "A novel optical imaging method for the early detection, quantitative grading, and mapping of cancerous and precancerous lesionsof cervix," IEEE Transactions on biomedical engineering, vol. 48, pp. 96-104, 2001.
[29]吴令英, et al., "醋酸涂抹法在宫颈癌筛查中的应用价值,"癌症, vol. 22, pp. 1096-1098, 2003.
[30]孙红, et al., "阴道镜在宫颈人乳头状瘤病毒感染诊断中的价值,"上海预防医学, vol. 18, pp. 11-12, 2006.
[31]杨伶俐, et al., "阴道镜下醋酸白色上皮的病理分析,"上海医学, vol. 30, pp. 104-106, 2007.
[32]柯然利,张爱俐, "阴道镜诊断宫颈HPV感染的价值,"实用临床医学(江西), vol. 6, pp. 82-82, 2005.
[33]向日华,王润生, "一种基于高斯混合模型的距离图像分割算法,"软件学报, vol. 14, pp. 1250-1257, 2003.
[34] J. Bilmes, "A gentle tutorial of the EM algorithm and its application to parameter estimation for Gaussian mixture and hidden Markov models," International Computer Science Institute, vol. 4, 1998.
[35] R. Redner and H. Walker, "Mixture densities, maximum likelihood and the EM algorithm," SIAM review, pp. 195-239, 1984.
[36] T. Lehmann and C. Palm, "Color line search for illuminant estimation in real-world scenes," Journal of the Optical Society of America A, vol. 18, pp. 2679-2691, 2001.
[37] H. Greenspan, et al., "A continuous probabilistic framework for image matching," Computer Vision and Image Understanding, vol. 84, pp. 384-406, 2001.
[38] G. Zimmerman-Moreno and H. Greenspan, "Automatic detection of specular reflections in uterine cervix images," 2006.
[39] R. S. Hunter, "abstract" Journal of the Optical Society of America vol. 38, p. 661, 1948.
[40] H. Greenspan, et al., "Automatic detection of anatomical landmarks in uterine cervix images," IEEE Transactions on medical imaging, vol. 28, 2009.
[41] R.冈萨雷斯著,阮秋琦译,数字图像处理:电子工业出版社, 2003.
[42]孙根云, et al., "图像的模糊边缘检测算法,"光电工程, vol. 34, pp. 141-144, 2007.
[43] Y. Wang, et al., "Facial feature detection and face recognition from 2D and 3D images," Pattern Recognition Letters, vol. 23, pp. 1191-1202, 2002.
[44] G. Pass and R. Zabih, "Comparing images using joint histograms," Multimedia systems, vol. 7, pp. 234-240, 1999.
[45] W. Cropper, "Rudolf Clausius and the road to entropy," American Journal of Physics, vol. 54, p. 1068, 1986.
[46] C. Burges, "A tutorial on support vector machines for pattern recognition," Data mining and knowledge discovery, vol. 2, pp. 121-167, 1998.
[47] V. Vapnik, The nature of statistical learning theory: Springer Verlag, 2000.