Evaluation of baseline wander removal approaches in ECG recordings

Cardiovascular Diseases (CVD) rank first among the causes mortality of the population of our planet according to the World health organization (WHO), accounting for about a third of all cases of death. In Europe, almost a million people die from these diseases every year (at least 50% of all deaths), of which about 100 thousand are people at working age. In addition, experts (WHO) forecast a further increase in cardiovascular morbidity and mortality, as in developed and developing countries, due to aging populations and lifestyle features. According to preliminary data, in 2030 myocardial infraction and stroke will remain the main individual causes of death. In this regard, there is a need for accurate and timely diagnosis of diseases of the cardiovascular system in the early stages development, which is one of the strategic directions of the health care industry. The basis of such a diagnosis is electrocardiography (ECG). This method is a non-invasive examination of the bioelectrical activity of the heart. Currently this method of diagnosis is indispensable for rhythm and conduction disorders, hypertrophy of the ventricles and atria, coronary heart disease, including painless form, myocardial infarction and other dangerous pathologies. However, in connection with the peculiarities of physiological origin and registration of signals, there is a serious and unsolved problem of their automatic analysis. ECG signals are unsteadily structured signals with cyclically repeating informative sections. Specifically, according to these signs, the state of the cardiovascular system is assessed in electrocardiography. Morphology and parameters are diverse, variable, and unpredictable. In addition, it should be borne in mind that, while electrocardiography is recorded, it is inevitable in record also numerous interferences and noises in the signal. There are various kinds and origins that influence signal, leading to morphological changes in the electrocardiogram. In some cases, such changes do not have diagnostic importance, but sometimes they can be mistakenly considered as signs of myocardial damage, leading to inaccurate or erroneous automatic conclusions. As a result of this problem, in cardiac diagnostics is development of various methods and algorithms for noise suppression. Have been developed each method for such problems has advantages and disadvantages, therefore, why in this project we analyze different methods to struggle against interferences. The effectiveness of noise suppression will be determined by the degree of suppression of interference and the degree of reservation of diagnostic information