Part of #Detection of Collagenous Colitis Based OnHistopathology Image Segmentation of Colon# :
Publishing year : 2011
Conference : Seventh Conference on Visual Machine and Image Processing
Number of pages : 5
Abstract: Collagenous colitis is prevalent among pathologicaldiseases. Because of normal radiological and endoscopicappearing, the microscopic image of colon tissue isnecessary and important material for the study of thisdisease. In this paper, a computerized system is developed forcollagenous colitis diagnosis. First, the colon tissues are coloredwith the Mason trichrome to make the sub-epithelial collagen band appear in blue. Second, the preprocessing is done on the chromosomal image of the colon tissue, and then the color histograms in the HSV color space and the Haar Wavelet energy indetail coefficients of the sub-band images are implemented to characterize the histological structure representation fortissue classification. The KNN Classifier is selected for this classification, which segment the sub-epithelial collagenband from the image, and finally, the thickness of the sub-epithelialcollagen band is determined by morphological operations. An abnormal case will be recognized if the thickness of the subepithelialcollagen band is greater than 10 m. Fifty clinical imaging samples were used in the training and testing procedures for the evaluation of system performance. Through the applied classification, it was revealed that the collagenous colitis can be accurately diagnosed.