Classification of aorta insufficiency and stenosis using MLP neural network and Neuro-fuzzy system

 

Fırat Hardalaç,1 Necaattin Barışçı,2 Uçman Ergün2

 

1. Department of Biomedical, Faculty of Medicine, Gazi University, Ankara (Turkey)

2. Department of Electronic and Computer Education, Faculty of Technical Education, Gazi University, Ankara (Turkey)

 

 

Manuscript received: April 15, 2004; revised: August 26, and October 27, 2004

Accepted for publication: November 18, 2004

 

Abstract

Cardiac Doppler signals recorded from aorta valve of 120 patients were transferred to a personal computer by using a 16 bit sound card. Spectral analyses of aorta valve Doppler signals were performed for determining the Multi Layer Perceptron (mlp) neural network and Neuro-fuzzy system inputs. In order to do a good interpretation and rapid diagnosis, ar parameters of aorta valve Doppler signals classified using mlp neural network and Neuro-fuzzy system. Our findings demonstrated that 90% correct classification rate was obtained from mlp neural network, and 88.33% correct classification rate was obtained from Neuro-fuzzy system. Since we had limited number of patient, there is no significant performance difference observed between the two methods.

 

keywords: Multi Layer Perceptron (mlp), Neuro-Fuzzy Classification (nefclass), Cardiac Doppler, Autoregressive (ar) spectral analysis.

 

 previous