Decision support system using multimedia case history quantitative comparison and multivariate statistical analysis

Abstract

A decision support system (DSS) for modeling and generalization of verified clinical information on individual patients is being developed. The data set for each patient is collected inform of multimedia case history (MMCH). The paper presents an approach to the data processing. The approach is based on case-to-case and case-to-cluster comparative and multivariate statistical analysis of the patients' data. Namely, the DSS use the normalization procedures in individual time-and-subspace domain and the interpolation techniques for multimedia data. It makes the individual MMCH data to be pair-wise comparable. Some evaluation results of the approach in area of hematology are presented. Sample set of relevant MMCH data has been obtained for a dozen of Chernobyl patients. The approach developed is to provide backgrounds for a unified language to meet needs of WHO in quantitative description, comparison and generalization of individual patients' patterns.