N.A. Al-khulaidi, R.V. Isakov, L.T. Sushkova
The paper presents results of research capabilities of automated classification of arrhythmias using neural network analysis of the correlation rhythmogram (scattergram).
Based on the scattergram, a database of arrhythmias was created, corresponding to the known types of heart diseases associated with cardiac arrhythmia. They were grouped into eight main classes.
Comparison of multi-layered perceptron and modular construction of neuron network organization was in-process produced, as variants of the automated analysis of arrhythmia.
Analysis of the results shows that the modular structure of neural network analyzer cardiac arrhythmias has the higher effectiveness.