
Scoring of radiation‐induced microuclei in cytokinesis‐blocked human lymphocytes by automated image analysis
Author(s) -
Verhaegen F.,
Vral A.,
Seuntjens J.,
Schipper N. W.,
de Ridder L.,
Thierens H.
Publication year - 1994
Publication title -
cytometry
Language(s) - English
Resource type - Journals
eISSN - 1097-0320
pISSN - 0196-4763
DOI - 10.1002/cyto.990170203
Subject(s) - micronucleus test , binucleated cells , false positive paradox , micronucleus , ionizing radiation , computer science , dosimetry , nuclear medicine , artificial intelligence , irradiation , physics , chemistry , medicine , toxicity , organic chemistry , nuclear physics
The micronucleus assay in human lymphocytes is, at present, frequently used to assess chromosomal damage caused by ionizing radiation or mutagens. Manual scoring of micronuclei (MN) by trained personnel is very time‐consuming, tiring work, and the results depend on subjective interpretation of scoring criteria. More objective scoring can be accomplished only if the test can be automated. Furthermore, an automated system allows scoring of large numbers of cells, thereby increasing the statistical significance for the results. This is of special importance for screening programs for low doses of chromosome‐damaging agents. In this paper, the first results of our effort to automate the micronuleus assay with an image‐analysis system are represented. The method we used is described in detail, and the results are compared to those of other groups. Our system is able to detect 88% of the binucleated lymphocytes on the slides. The procedure consists of a fully automated localization of binucleated cells and counting of the MN within these cells, followed by a simple and fast manual operation in which the false positives are removed. Preliminary measurements for blood samples irradiated with a dose of 1 Gy X‐rays indicate that the automated system can find 89% ± 12% of the micronuclei within the binucleated cells compared to a manual screening. © 1994 Wiley‐Liss, Inc.