МЕТОД СТАТИСТИЧНОГО ОЦІНЮВАННЯ РЕЗУЛЬТАТІВ КЛІНІЧНИХ ДОСЛІДЖЕНЬ ПРИ ЛІКУВАННІ АРТЕРІАЛЬНОЇ ГІПЕРТЕНЗІЇ ТА ОЖИРІННЯ

Abstract

Currently, the processing of a large amount of data, which have a different physical or biological nature, remains an important task, since all parameters have their own distribution laws, which in general differ from the Gaussian one. Such a set of data does not make it possible to apply standard methods of statistical processing. There is no general classificationmethod either.The analysis of research and publications made it possible to formulate the tasks of the research presented in the work, how to choose a relevant statistical approach that would optimally take into account the peculiarities of endocrinological research, which include anthropological, biochemical and other indicators. The paper proposes an approach based on image classification using distance functions.The purpose of the article is to compare methods of treating obesity and hypertension using statistical data processing methods, namely by calculating the Mahalanobis distance between groups of patients and to prove the effectiveness of the new proposed method of treatment.The article proposes a new approach to the statistical processing of a large amount of diverse data, based on the use of the Mahalanobis distance. The ways of decision-making based on the Mahalanobis distance measurement between a sample of relatively healthy patients and two samples of patients were considered: the main - a group of patients receiving a new treatment, and an experimental - a group of patients receiving standard treatment for excess weight and arterial hypertension.Therefore, the paper proposes for the first time the use of the Mahalanobis distance as a criterion for evaluating a large number of diverse physical indicators taken from a biological object.