Prediction of coronary angiography requirement of patients with Fuzzy Logic and Learning Vector Quantization

Abstract

In this study, prediction of coronary angiography (CA) requirement of patients is presented using Fuzzy Logic (FL) and Learning Vector Quantization (LVQ). Data sets of patients are received from 200 patients, half of whom undergo CA, the other half doesn't undergo CA, the numbers of both men and women patients are selected equal. Input data sets and output data sets are determined and tested for FL The correct classification rate of FL is measured 86% for prediction of CA requirement of patients. Training data sets and testing data sets are determined and tested for LVQ. The correct classification rate of LVQ is measured 88% for prediction of CA requirement of patients. These results show that LVQ is more effective than FL at prediction of CA requirement of patients.