Markovian analysis of cervical cell images.

Abstract

Markovian analysis is a method to measure optical texture based on gray-level transition probabilities in digitized images. Experiments are described that investigate the classification performance of parameters generated by Markovian analysis. Results using Markov texture parameters show that the selection of a Markov step size strongly affects classification error rates and the number of parameters required to achieve the maximum correct classification rates. Markov texture parameters are shown to achieve high rates of correct classification in discriminating images of normal from abnormal cervical cell nuclei. Texture is one of several properties (e.g., color, shape, size) by which human beings perceive images, and it is an important property for the classification of biomedical images. I This work was performed in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Biomedical Engineering,

Cytophotometric analysis of cervical intraepithelial neoplasia grade III, with and without synchronous invasive squamous cell carcinoma

“…Nuclear boundaries were automatically determined from the digital image by grey level boundary following. Twenty-one nuclear features, describing geometrical, density, and texture characteristics were extracted from the digitized image (Table 2) (7,8,10,20,21,24). As a n internal standard for DNA content, 20 lymphocytes were measured in each of the samples.…”

Brief communication: Deblurring and 3D-like rendering of light microscope images

“…Several quantitative methods have been developed to assess size, shape and density of cells and related structures from light microscope images (Colomb et al, 1991;Diaz et al, 1989;Diaz et al, 1996;Pressman, 1976). However, the numerical definition of structures has too often disregarded their visual demonstration.…”

DNA Ploidy and Markovian Analysis of Neoplastic Progression in Experimental Pancreatic Cancer

“…The nuclei of Feulgen-stained cells were evaluated for DNA ploidy and for a specific array of morphometrical characteristics, including 26 size, shape, and texture features, using a Cell Measurement Program (Cell Analysis Systems; Elmhurst, IL) combined with the CAS 200 (Pressman 1976;Bacus and Grace 1987;Dawson et al 1993). A total of 100-200 nuclei with clear boundaries appearing to have no loss of membrane integrity were identified for analysis from each tissue sample.…”

“…The remainder were placed in 10% buffered formalin for 24 hr and then prepared for light microscopy, deoxyribonucleic acid (DNA) analysis, and Markovian analysis using the CAS 200 computerized microscope (Becton Dickinson; Mountain View, CA) to determine surrogate endpoint biomarkers (Pressman 1976;Bacus and Grace 1987;Linder et al 1995;Dawson et al 1993). In all cases, tissue from the most abnormal-appearing portions of the pancreas was sampled.…”