Premium
Automatic quantification of viability in epithelial cell cultures by texture analysis
Author(s) -
Malpica N.,
Santos A.,
Tejedor A.,
Torres A.,
Castilla M.,
GarcíaBarreno P.,
Desco M.
Publication year - 2003
Publication title -
journal of microscopy
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.569
H-Index - 111
eISSN - 1365-2818
pISSN - 0022-2720
DOI - 10.1046/j.1365-2818.2003.01094.x
Subject(s) - linear discriminant analysis , artificial intelligence , texture (cosmology) , histogram , pattern recognition (psychology) , segmentation , phase contrast microscopy , microscopy , computer science , matrix (chemical analysis) , computer vision , mathematics , pathology , image (mathematics) , chemistry , medicine , physics , chromatography , optics
Summary Quantification of live cells in phase contrast microscopy images allows in vivo assessment of the viability of cultured cells. An automatic screening procedure seems advisable because of the large number of cells that must be counted to achieve reasonable accuracy. This paper presents a method that quantifies necrosis in cell cultures by texture analysis of microscope images. The image is divided into regions of equal size that are classified by means of a segmentation algorithm based on texture analysis into three categories: live cells, necrotic cells and background. The classification uses three discriminant functions, built from parameters derived from the histogram and the co‐occurrence matrix and calculated by performing an initial stepwise discriminant analysis on 21 sample images from a training set. The areas occupied by live and necrotic cells and number of live cells have been obtained for primary cellular cultures in intervals of 48 h during 2 weeks. The results have been compared with those obtained by an experienced observer, showing a very good correlation (Pearson's coefficient 0.95, kappa 0.87, N = 1600). A method has been developed that provides an accuracy similar to that provided by an expert, while allowing a much higher number of fields to be counted.