The primary goal is to develop and utilize quantitative high resolution image analyses techniques for the analysis and classification of blood samples from sickle cell patients. Specific tasks include: 1) refining classification features and algorithms through the creation of an image data base containing cell images together with specific clinical information and a consensus diagnosis on each cell; 2) developing an automated instrument for classifying blood cells into categories of normal (discocyte), abnormal (poikilocyte) and sickle, as defined by human observers; 3) studying subpopulations of normal and sickle cells identified through clustering, partitioning and multivariate graphical techniques; 4) studying cell features, subpopulations, and cell classification from specimen aliquots obtained by fractionation techniques and from cells exposed to various levels of oxygenation and deoxygenation; 5) studying and correlating quantitative cellular features from specimens obtained longitudinally during the course of randomized clinical trials with clinical course and outcome as provided by standardized assessment measures to seek quantitative features providing clinically useful information; 6) utilizing quantitative features providing to study heterogeneity of disease and heterogeneity of response to therapy; 7) comparing cellular features from SS with those from other double heterozygous forms of sickle cell disease: SC, SB+ Thalassemia, SBo Thalassemia; 8) assisting in the application of quantitative image analysis techniques in screening potential new antisickling agents; and 9) studying slit-scan light scatter to determine the utility of this technique in the analysis of sickle cells. Specimens for all studies will be obtained during the course of a randomized clinical trial at three institutions, and from the patient populations attending two sickle cell centers. The results of these studies should enhance the monitoring of patients in clinical trials, the testing of large numbers of potential antisickling agents, the analysis of drug effects, and provide information of clinical utility.
| Wheeless, L L; Robinson, R D; Lapets, O P et al. (1994) Classification of red blood cells as normal, sickle, or other abnormal, using a single image analysis feature. Cytometry 17:159-66 |
| Robinson, R D; Benjamin, L J; Cosgriff, J M et al. (1994) Textural differences between AA and SS blood specimens as detected by image analysis. Cytometry 17:167-72 |
