Granulometric analysis of corneal endothelium specular images by using a germ-grain model

Authors:
M. E. Díaz;G. Ayala;R. Sebastian;L. Martínez-Costa
Affiliations:
Departamento de Informática, Universidad de Valencia, Avda. Vicente Andrés Estellés, s/n, 46100 Burjasot, Spain;Departamento de Estadística e Investigación Operativa, Universidad de Valencia, Avda. Vicente Andrés Estellés, s/n, 46100 Burjasot, Spain;Departamento de Informática, Universidad de Valencia, Avda. Vicente Andrés Estellés, s/n, 46100 Burjasot, Spain;Hospital Doctor Peset, Avda. Gaspar Aguilar, 90, Valencia, Spain
Venue:
Computers in Biology and Medicine
Year:
2007

Citing 4
Cited 0

Generalization of shift invariant neural networks: image processing of corneal endothelium

Neural Networks
Spatial Size Distributions: Applications to Shape and Texture Analysis

IEEE Transactions on Pattern Analysis and Machine Intelligence
Set descriptors for visual evaluation of human corneal endothelia

Computer Vision and Image Understanding
Morphological Image Analysis: Principles and Applications

Morphological Image Analysis: Principles and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

Specular microscopy is widely used to study the human corneal endothelium status in vivo. In this paper, the corneal endothelium is represented as a binary image composed of the cell inscribed circles. The granulometric distribution function of the complement of this image is used as a functional descriptor, which provides information about the shape, size and spatial arrangement of cells. Experimental evaluation using bootstrap techniques shows its ability to discriminate between controls and pathological cases. It represents a reliable and graphical alternative to the classical indices (cell density, hexagonality and coefficient of variation of cell areas), which behave poorly when detecting subtle abnormalities.