GENERATION OF SYNTHETICAL MEDICAL DATA BY MDR-ANALYSIS

Purpose: The purpose of this article is to outline an algorithm for generating synthetic medical data in order augment small samples data. Methods: To achieve the research goal, methods such as: correlation analysis (to identify significant variables and relationships between them), MDR build logical chains data), regression model use generate data) were used. Results: A database heart failure patients that publicly available was used test developed action; as a result, statistical found linear models. Discussion: proposed allows, with few simple, yet important actions, perform generation data, which makes it possible obtain large sets can be implement machine learning any tasks related medicine.