CREATING MODELS FOR IMPROVING DIABETES THERAPY WITH MULTIVARIATE DATA VISUALIZATION

Savoska, Snezana and Savoski, Z. and Manevska, Violeta (2016) CREATING MODELS FOR IMPROVING DIABETES THERAPY WITH MULTIVARIATE DATA VISUALIZATION. In: Interdiab conference 2016, 3-5 March 2016, Bucharest, Romania.

Full text not available from this repository.
Official URL: https://www.researchgate.net/profile/S_Savoska/pub...

Abstract

Methods for multivariate visualization can be used to visualize diabetes patient's data as well as to improve visual insight for interdependences between variables analyzed by endocrinologists. These insights can be used in defining the data clusters of patients with different behavior. Depending on the patients' behavior, according to previously defined independent variables (insulin therapy, diet-bread units and physical activity-aerobics exercise) can create models of expectations to achieve the dependent variables (level of HBA1c, bodyweight and patient satisfaction). Gained models can be used in medical expert systems as well as to enrich medical knowledge base systems. Multivariate methods of data visualization for patients with diabetes, for endocrinologists, means getting original visual perception, behavior of clusters of patients that will be used as a basis for modeling different patients' behavior-tuning independent variables to achieve desired values for dependent variables.

Item Type: Conference or Workshop Item (Paper)
Subjects: Scientific Fields (Frascati) > Medical and Helath Sciences > Clinical medicine
Scientific Fields (Frascati) > Engineering and Technology > Electrical engineering, electronic engineering,information engineering
Divisions: Faculty of Information and Communication Technologies
Depositing User: Prof. d-r. Snezana Savoska
Date Deposited: 18 Jun 2020 13:44
Last Modified: 18 Jun 2020 13:44
URI: http://eprints.uklo.edu.mk/id/eprint/5457

Actions (login required)

View Item View Item