Automatic Detection of Vaginal Bacteria Based on Superpixel and Support Vector Machine
1 Department of Biomedical Engineering，School of Medicine, Shenzhen University, Guangdong Key Laboratory for Biomedical Measurements and Ultrasound Imaging, Shenzhen 518060, Guangdong, China
2 Department of Laboratory, the Sixth People’s Hospital of Shenzhen, the Affiliated Hospital of Nanshan, Shenzhen University, Shenzhen 518052, Guangdong, China
Abstract The vaginal disorders or infections caused by bacteria may lead to serious diseases including ectopic pregnancy, infertility, acute or chronic pelvic inflammatory. In clinics, cytology screening is used to prevent vaginal disease; however, misdiagnosis and missed diagnosis occasionally occurred. Here we presented an algorithm for the automatic detection of vaginal bacteria based on the superpixel and support vector machine using images from Gram stain. The algorithm applied simple linear iterative clustering to compute superpixel. We characterized these superpixels areas using features of shape and color and histograms of oriented gradients, and then performed support vector machine classification.〓Fourty negative of BV and 60 indicative of BV images were chosen, one of 10 negative of BV and 20 indicative of BV images were used for training the classifier, the rest of the images for testing the algorithm. Results suggest that the proposed method reached the average ZSI of 8927%, showing promising potentials of clinical application.
［1］Forsum U, Holst E, Larsson PG, et al. Bacterial vaginosisa microbiological and immunological enigma［J］. Apmis, 2005, 113: 81-90.
［2］Donders D. Diagnosis and management of bacterial vaginosis and other types of abnormal vaginal bacterial flora: a review ［J］. Obstetrical & Gynecological Survey, 2010, 65: 462-473.
［3］Turnbaugh PJ, Ley RE, Hamady M, et al. The human microbiome project: exploring the microbial part of ourselves in a changing world ［J］. Nature, 2007, 449(7164):804.
［4］Nugent RP, Krohn MA, Hillier S. Reliability of diagnosing bacterial vaginosis is improved by a standardized method of gram stain interpretation ［J］. Journal of Clinical Microbiology, 1991, 29:297-301.
［5］Forero M, Cristóbal G, Desco M. Automatic identification of Mycobacterium tuberculosis by Gaussian mixture models ［J］. Journal of Microscopy, 2006, 223:120-132.
［6］Chang J, Arbeláez P, Switz N, et al. Automated tuberculosis diagnosis using fluorescence images from a mobile microscope［C］ //Nicholas A, eds. 15th International Conference on Medical Image Computing and ComputerAssisted Intervention. Berlin:Springer, 2012:345-352.
［7］孙涵璇, 谢凤英, 姜志国, 等. 基于 BP 神经网络的结核杆菌目标识别［J］. 中国体视学与图像分析, 2010, 1:13-17.
［8］翟永平, 周东翔, 刘云辉. 基于颜色及梯度统计特征的结核杆标识别［J］. 国防科技大学学报, 2012, 34:146-152.
［9］Comaniciu D, Meer P. Mean shift: A robust approach toward feature space analysis［J］. IEEE Transactions on Pattern Anal Mach Intell, 2002, 24:603-619.
［10］Vedaldi A, Soatto S. Quick shift and kernel methods for mode Seeking［C］.//Jean P, eds. 10th European Conference on Computer Vision. Berlin: Springer, 2008: 705-718.
［11］Levinshtein A, Stere A, Kutulakos KN, et al. Turbopixels: Fast superpixels using geometric flows［J］. IEEE Transactions on Pattern Anal Mach Intell, 2009, 31:2290-2297.［12］Shi J, Malik J. Normalized cuts and image segmentation［J］. IEEE Transactions on Pattern Anal Mach Intell, 2000, 22:888-905.
［13］Tighe J, Lazebnik S. Superparsing: scalable nonparametric image parsing with superpixels［C］.//Argyors A, Trahanias P, Tziritas G, eds. 11th European Conference on Computer Vision. Berlin: Springer, 2010: 352-365.
［14］Gorelick L, Veksler O, Gaed M, et al. Prostate histopathology: learning tissue component histograms for cancer detection and classification［J］. IEEE Transactions on Med Imag, 2013, 32:1804-1818.
［15］Achanta R, Shaji A, Smith K, et al. SLIC superpixels compared to stateoftheart superpixel methods［J］. IEEE Transactions on Pattern Anal Mach Intell, 2012, 34:2274-2282.
［16］Dalal N, Triggs B. Histograms of oriented gradients for human Detection［C］.//Martial H, David K, eds. IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York: IEEE Society, 2005:886-893.
［17］Cortes C, Vapnik V. Supportvector networks ［J］. Machine Learning, 1995, 20:273-297.
［18］Burges CJ. A tutorial on support vector machines for pattern Recognition［J］. Data Mining and Knowledge Discovery, 1998, 2(2):121-167.
