Texture analysis of fluorescence microscopic images of colonic tissue sections

Abstract

The aim of this study was to assess the potential of texture analysis for the characterization of fluorescence images from colonic tissue sections stained with a novel and selective fluoroprobe, Rhodamine B-phenylboronic acid. Fluorescence microscopy images of colonic healthy mucosa (n = 35) and adenocarcinomas (n = 35) were digitally captured and subjected to image texture analysis. Textural features derived from the grey level co-occurrence matrix were calculated. A modified version of the multiple discriminant analysis criterion was used to choose an appropriate subset of features. A minimum Mahalanobis distance, linear discriminant classifier and a simple evaluation 'score' method were used to classify image feature data into the two categories. A subset of four textural features was selected and used for the description and classification of each image field. They were found appropriate to correctly classify 95% of the images into the two classes, using two different classifiers. These features contained information about local homogeneity and grey level linear dependencies of the image. This study demonstrated that texture analysis techniques could provide valuable diagnostic decision support in a complex domain such as colorectal tissue.

Texture analysis for tissue classification of optical coherence tomography images

“…Texture analysis has also been used to evaluate ultrasound images of the prostate [18]. Other optical imaging modalities have utilized texture analysis, such as fluorescence microscopy images of colonic tissue sections [19] and light microscopy images of the chromatin structure in advanced prostrate cancer [20].…”

Abdominal Tumor Characterization and Recognition Using Superior-Order Cooccurrence Matrices, Based on Ultrasound Images

“…They arise from the adenomatous polyps which are present on the bowel wall. Like every tumor, they are characterized by the heterogeneity of the tissue structure and by the complexity and irregularity of the vessel structure [5]. In ultrasound images, they have an inhomogeneous, mixed aspect, the parietal delimitation being linear, but interrupted by the tumor invasion.…”

“…The method provided satisfying results concerning the differentiation between malignant and benign liver lesions, the area under the ROC (receiver operating characteristic) curve being approximately 90%. In [5] the authors analyzed the fluorescent images of the colonic tissue based on textural parameters derived from the second order grey level cooccurrence matrix (GLCM), in order to distinguish the colonic healthy mucosa versus adenocarcinoma. However, a systematic study concerning the most relevant textural features that best characterize the malignant tumors and of the most appropriate methods that lead to an increased diagnosis accuracy is not done.…”

Implementation of GA-KSOM and ANFIS in the classification of colonic histopathological images

“…For instance in [7], entropy and correlation from GLCM were obtained from normal and cancerous colonic images and used with linear discriminant function and k-nearest neighbor classifiers. GLCMs were also used in [8] to extract features from a total of 70 fluorescence microscope images of colonic tissue sections stained with a novel selective fluoroprobe. Directional GLCMs for each image were combined into a non-directional GLCM by averaging values from four angular directions [0 o , 45 o , 90 o , 135 o ] separated by a distance of 1 pixel.…”