Search for hidden patterns affecting overall patient survival with data mining

The reasons affecting the survival time of patients with chronic lymphocytic leukemia, taking into account gender, are studied. The set of patients is divided into two disjoint subsets (classes) by the indicator of actual survival, the value of which is less than the predicted value of overall survival, is determined. To detect hidden patterns in the analysis, nonlinear data transformations based on the calculation of the values of the class membership function for each attribute are used. The threshold values between the classes on the numerical axis are determined, both by individual attributes and by generalized assessments of objects on defined sets of attributes. The threshold values are used to record logical patterns in the form of half-planes and display gender differences for predicting the survival of patients.