학술논문

Using data mining methods for identification relationships between medical parameters
Document Type
Conference
Source
2017 IEEE 21st International Conference on Intelligent Engineering Systems (INES) Intelligent Engineering Systems (INES), 2017 IEEE 21st International Conference on. :000049-000054 Oct, 2017
Subject
Bioengineering
Communication, Networking and Broadcast Technologies
Computing and Processing
Engineering Profession
Robotics and Control Systems
Data mining
Diseases
Medical diagnostic imaging
Heart
Histograms
Tools
Prognostics and health management
Language
Abstract
The aim of this article is to analyze the medical data using the data mining methods Under medical data or biomedical data in this article, we understand data that describe the health state of patients with the diagnosis of ischemic heart disease on the basis of results of underwent examinations. At the same time, we designed a suitable way of applying the data mining process when analyzing this data. The entire process is divided into several parts. The first part is devoted to the general identification of problems of acquiring knowledge from medical data, such as identification of the diagnosis or multiple diagnoses of the patient, or identification of the effect of medical parameters on the outcome of the patient's prognosis. The next part is focused on identifying and collecting medical data for the purpose of discovering new knowledge. During this phase, the method for medical data collection from hospital reports was designed. These reports indicated the patient's health status during the hospitalization or examination period. In the data mining phase, medical data was analyzed using selected data mining methods. At this stage, the degree of impact of individual parameters on the final prognosis was also determined.