Increase of precision and accuracy of DNA cytometry by correcting diffraction and glare errors.

In recent TV-based image cytometers considerable disproportionalities exist between the IOD values of reference cells, as well as diploid, tetraploid, and octoploid analysis cells compared with their theoretical IOD ratios. An important source of these deviations is the limited spatial resolution of the microscopic objectives, based on the effects of diffraction. Compared to these influences the glare is less important. A correcting method is given for reducing both effects on the DNA measurements, which considers a narrow region along the nuclear contour to be optically disturbed. The correction of this mean optical density (MOD)- and size-dependent geometric resolution error is applicable to any cell type. The method was tested on 25 rat liver imprints and 29 fine needle aspirates from breast cancers. The resulting stemline ratios are close to the theoretical ones. A further improvement was then reached by a glare correction. The slide-by-slide variations of stemline ratios, remaining after the corrections, were considered in a test statistic for defining the aneuploidy of a given stemline. Statistically based clear-cut decision rules were obtained for DNA histogram interpretation.