SummaryChromatin distribution re¯ects the organization of the DNA of a nucleus and contains important cellular diagnostic and prognostic information. Feulgen staining of breast tissue enables the chromatin distribution of the nucleus to be visualized in the form of texture. Describing texture in an objective and quantitative way by means of a set of texture parameters, combined with the study of the relationship of such parameters to the pathobiological cell properties, is useful both for reduction of the subjectivity inherently coupled to visual observation and for more accurate prognosis or diagnosis.We have presented an automated classi®cation scheme for the diagnosis and grading of invasive breast cancer. The input to this scheme was a digitized microscopical image, from which nuclei were segmented. Chromatin texture was described using a set of textural parameters that include ®rst-and second-order statistics of the image grey levels. The more recently developed wavelet energy parameters were also included in our study. Classi®cation was performed by a Knn-classi®er, which is a versatile multivariate statistical classi®cation technique.We investigated the role of the tissue preparation technique and found that parameters derived from cytospins were better texture descriptors than those from sections. A 100% correct classi®cation was achieved in a patient diagnosis experiment and 82% in a nuclear grading experiment.

The influence of fixation, air-drying and liquid-imaging on the morphology as well as on the viscoelasticity of malignant mesothelioma cells was studied by atomic force microscopy. In this study, dehydrated cells were more easily scanned and offered faster data recording than hydrated cells. However, the influence of fixation strength was more noticeable. Strong fixation induced flattening of the cytoplasm and loss of nuclear structure, resulting in a clearly visible cytoskeleton which could be easily seen as fibres orientated in the direction of the cell growth. By contrast, the morphology of hydrated cells was influenced to a lesser degree on fixation and showed an overall 'rounding' of the surface with vague, ill-defined structures. Nuclear areas of these samples were difficult to image. Viscoelasticity measurements also exhibited large differences. Dehydrated cells were much harder and showed a uniform indentation profile over the whole cell that was independent of fixation. Indentation on hydrated cells was large and depended on the height of the measuring spot, the submembranous structure and, to a lesser extent, on fixation. To calculate an overall 'cellular' viscoelasticity, different methods were tested on these samples. Indentations of multiple, randomly chosen points, covering the whole cell, were measured and averaged to yield a mean indentation score. We avoided the thin and shadowed areas since it was shown that these regions were less suited for measuring. Using this design, large viscoelasticity differences were found, on which the influence of the external parameters could be shown. In another set-up, layered imaging was tried. However, long data acquisition times caused cellular activation and rearrangement, making this scanning mode unsatisfactory.

No abstract