Abstract:
The paper considers cardiac arrhythmias analysis methods starting with electrocardiogram (ECG) signal splitting into cycles, and finishing with the formation of the arrhy...Show MoreMetadata
Abstract:
The paper considers cardiac arrhythmias analysis methods starting with electrocardiogram (ECG) signal splitting into cycles, and finishing with the formation of the arrhythmias types sequence. The input data for the analysis are the results of splitting the ECG-signal into cycles. It is suggested to use a modification of Cobweb method for clustering of QRS-complexes of ECG-signals. Formation of arrhythmias dictionary based on sequential analysis (n-gram method) of ECG rhythm in the text form is described. Linguistic analysis for arrhythmia type classification is proposed. The examples of analysis of the ECG-signal record from MIT-BIH arrhythmia database are given. The proposed methods were tested in an integration mode with Holter system.
Date of Conference: 20-24 September 2021
Date Added to IEEE Xplore: 24 December 2021
ISBN Information: