Data processing in electrocardiographs by wavelet transformation for early forecasting of parossysmal arthritis

Abstract

ECG analysis is widely used to diagnose many heart diseases, which are the leading cause of death
in different countries. The quality of the ECG signal can be affected and worsened by various
sources, such as the patient’s condition, basic walk, electrocardiogram contact, and others. In
addition, if the ECG is visually monitored, the probability of human error is high, each 10-result
is interpreted with an error (Brikena Xhaja, 2015: 305-312) And also for many ECG images it is
simply not possible to conduct a visual analysis of the frequency data of the signal. The morphology
of low-amplitude high-frequency signals, the so-called P waves, hides valuable information for early
preclinical disease prediction. That is, the need to search for new methods of early preclinical
diagnosis is still relevant. Since the majority of clinically useful information in the ECG is found
in the intervals and amplitudes determined by its significant points (characteristic peaks and wave
boundaries), the development of accurate and reliable methods for automatic ECG delineation is
a matter of great importance, especially for the analysis of long records (Juan Pablo Martinez,
2014: 570-581). The problems of retrieving information from the electrophysiological signal that
can not be obtained by visual analysis of the recording, as well as the problems of automation
of traditional algorithms of medical analysis are relevant in connection with the lack of research
in this field. The aim of the research is to search for new areas of application of the wavelet
transform method in signal processing. Wavelet transformation, obtained widely in 2000 in the
study of signal properties, allows us to "discern"hidden frequency-time signal data with the help of
approximating and detailing coefficients. The obtained results show that the proposed algorithm
provides real efficiency in the processing of primary signals for the task of isolating the detailing
coefficients of the ECG signal. Our study shows that Morlet’s wavelet analysis of P intervals, which
can be applied easily and inexpensively, can reliably predict the incidence of symptomatic episodes
of paroxysmal atrial fibrillation in patients without clinically and echocardiographically expressed
heart disease. Wavelet analysis can contribute to our understanding of the electrophysiological
mechanisms underlying the generation and recurrence of paroxysmal atrial fibrillation and can
identify patients at high risk of increased relapses of paroxysmal atrial fibrillation, thereby creating
the prospect of early application of non-invasive and invasive therapeutic strategies to prevent
future events of paroxysmal ciliary arrhythmias.