Risk Prediction of Ischemic Heart Disease Using Artificial Neural Network

<p>The assessment model of coronary heart disease is so much developed in line with the development of information technology, particularly the field of artificial intelligence. Unfortunately, the assessment models developed mostly do not use such an approach made by the clinician, the tiered approach. This study aims to analyze the performance of a tiered model assessment. The method used for each level is, preprocessing, building architecture artificial neural network (ANN), conduct training using the Levenberg-Marquardt algorithm and one step secant, as well as testing the system. The study is divided into the terms of the stages in the examination procedure. The test results showed the influence of each level, both when the output level of the previous positive or negative, were tested back at the next level. The performance evaluation may indicate that the top level provides performance improvement and or reinforce the previous level. </p>

Download Full-text

Analysis of evaluation problems of the risk situation of patieents suffering from ischemic heart disease

Acta Universitatis Agriculturae et Silviculturae Mendelianae Brunensis ◽

10.11118/actaun201260020125 ◽

2012 ◽

Vol 60 (2) ◽

pp. 125-134 ◽

Cited By ~ 2

Author(s):

Vladimír Konečný ◽

Milan Sepši ◽

Oldřich Trenz

Keyword(s):

Neural Network ◽

Heart Disease ◽

Ischemic Heart Disease ◽

Complex Analysis ◽

Partial Solution ◽

Health Condition ◽

Ischemic Heart ◽

Classification Model ◽

Data Set ◽

Software Application

The ischemic heart disease represents a very common health issue which, thanks to its seriousness, impacts a big part of the population and is the cause of about one third of all death cases in the Czech Republic. For the analysis itself, data from medicinal practice of one of the authors of the article have been used and this study is a follow up of his PhD thesis. Concretely it was a set of patients which were being rehabilitated after a heart stroke; the results of the medical examination of these patients create 26 parameters. This data has been obtained in the course of the patients’ treatment. In the first phase of generating the classification model, the parameters that didn’t have a detrimental effect on the assessment of health condition of the patients have been removed from the data set and have been kept in the category of additional parameters. For the classification itself, an approach from artificial intelligence – applying a neural network - has been chosen. For the recording and transformation of the entering data a special application has been made. The classification and analysis of the data is performed on an experimental model of the self-learning of a neural network. The conclusions that arise from the initial analysis of this issue and the partial solution can be generalized and when using an appropriate software application they could even be used in medical practice. To do a complex analysis of the influence of all 26 parameters on the overall state of health of the patients is very difficult. A decision-making model appears to be a good solution. Last but not least, the proposed solution has to be verified on a bigger sample of patients afflicted by the ischemic heart disease.

Download Full-text