The present study deals with the application of two major multivariate statistical approaches - Cluster Analysis (CA) and Principal Components Analysis (PCA) as an option for assessment of clinical data from diabetes mellitus type 2 patients. One hundred clinical cases of patients are considered as object of the statistical classification and modeling, each one of them characterized by 34 various clinical parameters. The goal of the study was to find patterns of similarity, both between the patients and the clinical tests. Each group of similarity is interpreted revealing at least five clusters of correlated parameters or five latent factors, which determine the data structure. Relevant explanation of the clustering is found based on the pattern of similarity like glucose level, anthropometric data, enzyme level, liver function, kidney function etc. It is assumed that this classification could be of help in optimizing the performance of clinical test for this type of patients and for designing a pattern for the role of the different groups of test in determining the metabolic syndrome of the patients.
Financed by the National Centre for Research and Development under grant No. SP/I/1/77065/10 by the strategic scientific research and experimental development program:
SYNAT - “Interdisciplinary System for Interactive Scientific and Scientific-Technical Information”.