CLASSIF1 Data Pattern Classification

• < Classification of Risk Patients

- The print output of the CLASSIF1 procedure CLSTHROM comprises the classification report and information on the parameters of the disease masks. The disease masks (fig.3). are determined by iteratively identifying the most discriminatory database columns (n=5) of the 44 data columns from the IgG/, CD62/, CD63/ and thrombospondin/FSC/SSC measurements (fig.2)

- The most informative database columns derive from all four measurements. The IgG/CD62/CD63/thrombospondin measurements contribute 1/1/1/2 data columns. The selected database columns are all antibody intensity or surface density values. Percent frequency values of antibody negative or positive cells, in contrast, were not chosen because they were less discriminatory.

Standardized Classification Report

Information on Learning Set and Classification Parameters

• < Classification of Risk Patients