Modification of the method of optimal valid partitioning for comparison of patterns related to the occurrence of ischemic stroke in two groups of patients

Abstract

The paper presents new modifications of a previously developed method of optimal valid partitioning (OVP), including a new way of verifying two-dimensional patterns, which allows statistically valid estimation of the contribution into the pattern of each of the relevant variables. The means for analyzing the systems of patterns found by using the OVP method are also presented. A new variant of the OVP method, which allows searching for statistically valid differences between the patterns that exist in two groups of objects, is considered. An example of the practical use of the OVP versions in biomedical research is given. The differences between the risk factors for ischemic stroke in the group of patients who had transient ischemic attack (TIA) and in the group of patients without TIA are revealed.