The Cytometry and Biostatistics Core provides expertise and resources to individual projects and other cores in: 1) Cytometric analysis of all types of human breast material used in this program; 2) Database management for archiving, organizing, and accessing all patient information and scientific data generated by this program; and 3) Biostatistical analysis of our clinical data and scientific results to generate models of human breast cancer that more accurately predict biological behavior and patient diagnosis, prognosis, and response to therapy than do current approaches. Cytometry primarily is slide-based and encompasses the following: 1) Viability Index, by fast green exclusion; 2) DNA Cytometry, to give DNA Index and ploidy profiles; 3) Morphometry, particularly of nuclear morphology and patterns of chromatin distribution; 4) Cytokinetics, by incorporation of bromodeoxyuridine or expression of nuclear proliferative antigens; 5) Quantitative Immunocytochemistry, by labeled monoclonal or polyclonal antibodies; 6) Molecular Cytochemistry, by in situ hybridization with labeled probes. Our master database is UNIX based, but will be accessed for data entry, maintenance, and reporting by the individual investigators through personal computers that run commercial software and are linked with one another and the master database by telephone. Biostatistical Analysis is based on multiple approaches to evaluate our data and to develop optimal models of breast malignancy, including linear stepwise discriminant, multivariate regression, Cox multifactorial, classification and regression trees, progressive least squares, and cluster analyses.
Showing the most recent 10 out of 133 publications
