MATHEMATICAL CLASSIFICATION MODELS FOR MULTI-CHANNEL SIGNALS OF THE ELECTRIC FIELD OF A HEART
I. P. Polyakova, T. B. Feofanova, N. V. Luk'yanova VANT. Ser.: Mat. Mod. Fiz. Proc 2011. Вып.4. С. 82-88.
Neuronet models for decreasing the size of the data array obtained at multi-channel registration of electrocardiogram are discussed. The model based on the self-organization algorithm is used for the detection of meaningful features in the input data array. The proposed models were tested on the determination of the anomalous impulse paths in patients' hearts suffering of the preexcitation syndrome.Keywords: neuronets, generalized Hebb algorithm, self-organization maps, multi-channel electrocardiogram signals.
|