Title :
Inter-extrema profile analysis for a texture coarseness estimate in automatic metaphase finding
Author :
Vega-Alvarado, Leticia ; Márquez, Jorge ; Corkidi, Gabriel
Author_Institution :
Centro de Instrumentos, Univ. Autonoma Metropolitana, Mexico City, Mexico
Abstract :
As part of a larger effort for automated metaphase finding in cell proliferation studies, we report our experiences with a texture coarseness measure of intensity profile samples. This measure is highly specific to gray-level inter-extrema roughness features in microscopic images of metaphase spreads. Chromosome fragments produce patterns of pixels at low-resolution, and the local neighborhood of their individual extrema presents a characteristic coarseness. Artifacts as well as other cytological objects like stimulated or non-stimulated nuclei give rise to similar discrete patterns. Using a stereological-based approach, we propose the inter-extrema profile range (IEPR) analysis, which allows the estimation of 2D texture coarseness through statistical features of 1D signatures. Results of its use on images of fields of metaphases and artifacts are compared with the mean depth-width ratio of extrema and other texture measures
Keywords :
cellular biophysics; feature extraction; image texture; parameter estimation; 1D signature statistical features; 2D feature extraction; 2D texture coarseness estimate; automatic metaphase finding; cell proliferation studies; chromosome fragments; cytological objects; gray-level inter-extrema roughness features; inter-extrema profile analysis; inter-extrema profile range analysis; mean depth-width ratio of extrema; metaphase spreads; microscopic images; stereological-based approach; Automation; Bayesian methods; Biological cells; Data mining; Image texture analysis; Instruments; Microscopy; Neural networks; Surface morphology; Testing;
Conference_Titel :
Image and Signal Processing and Analysis, 2001. ISPA 2001. Proceedings of the 2nd International Symposium on
Conference_Location :
Pula
Print_ISBN :
953-96769-4-0
DOI :
10.1109/ISPA.2001.938621