Data Mining and Endocrine Diseases: A New Way to Classify?

Abstract

Data mining consists of using large database analysis to detect patterns, relationships and models in order to describe (or even predict) the appearance of a future event; to accomplish this, it uses classification methods, rules of association, regression patterns, link and cluster analyses. Recently this approach has been used to propose a new diabetes mellitus classification, using information analysis techniques through which the selection bias minimally influences categorization, this new focus that includes data mining previously implemented to predict, identify biomarkers, complications, therapies, health policies, genetic and environmental effects of this disease; it could be generalized in the field of endocrinology, in the classification of other endocrine diseases.