Automated selection of metaphase cells by quality,

Abstract

The performance of metaphase-finding systems could be improved if they were able to determine the quality of the cells detected. This paper discusses the extent to which this may be realized by the introduction of a metaphase-quality parameter. Data obtained from 300 cells were statistically analyzed. Seventeen features were measured and nine visual properties were determined for each cell. Discriminant analysis and regression analysis were used to extract those features and visual properties which contribute to assessment of metaphase quality. Rather low correlations were found beIn the early 1970s several groups worked on systems for automated chromosome analysis (2, 3, 10, 12, 13). At f i s t , the ambitious aim was to develop fully automatic analysis systems; later an approach with operator-interaction seemed to be more appropriate. However, interest became focussed more and more on automating the localization of metaphase cells on the microscope slide. It was thought that, in a two-step approach, a system should first be able to search in a fully automatic way for suitable cells and store their slide coordinates, and that later, in a semiautomatic mode, the cells which had been located would be quickly recalled one by one and examined as to their suitability for further analysis (4,5,7, 13, 20, 21).Until now, most of the systems developed for automated metaphase selection have been designed to be fast enough to meet arbitrary criteria for an acceptable operating speed. It appears that, for many researchers, operating speed was a more important factor than the quality of the found metaphases.The importance of determining the quality of the automatically selected cells is mentioned in previous publications, and such an approach is seen as a promising way to improve the ' Presented at Automated Cytology VIT, Asilomar, California, November 25-30,1979, Supported by "Praeventiefonds", Grant 2826411.tween the selected measured features and visual properties. A quality-parameter based on a linear combination of cluster projections, areas and perimeters was found to account for 64% of the variation between visual and measured quality indicators. In addition, an increase in the predictive value for finding usable metaphases from 2 8 4 8 % was achieved.

Radiation Dose Estimation by Completely Automated Interpretation of the Dicentric Chromosome Assay

“…Rule-based criteria for metaphase assessment have also extracted morphological features of objects enclosed by rectangles to partition chromosomes (7) or images (8) into small numbers of generally analyzable classes, with the exception of poor quality samples with high levels of debris or low mitotic indices (8) . User-scored features have been combined into quality measures which then classify metaphase cells into discrete groups (9) . Our approach integrates aspects of metaphase cell selection with radiation dose determination.…”

MetaSel: a metaphase selection tool using a Gaussian-based classification technique

“…There is thus a need for a more thorough and efficient method of selecting good metaphase spreads for karyotyping. Although some techniques have been proposed for automatic metaphase selection, in practice these techniques are impractical for processing hundreds of images in a typical cytogenetic analysis owing to the high computational complexity [1-5]; [13-15]. …”