Identification of cardiac arrhythmias by means of Wavelet Packet-based features
Cardiac arrhythmias are important indicators of heart diseases, in particular, they refer to electrical conduction problems and therefore their diagnosis and detection is of high clinical interest. However, in many cases, appropriate and timely detection is not very feasible due to several factors, such as computational cost, large amount of heartbeats per record, morphology variability, infrequency and irregularity of pathological heartbeats, among others. In this work, wavelet transform computed through wavelet packets is applied over electrocardiographic (ECG) signals as a method to characterize and identify normal ECG signals and some arrhythmias such as atrial fibrilation (AF) and life threaded arrhythmias, drawn from MIT databases. As a result, it is obtained a feature set that consist of wavelet relative energy and its standard deviation computed over the first decomposition levels. Characterization is assessed by means of wavelet trees which show a separability between normal and pathological segments of ECG signal due to timefrequency information generated by wavelet packets that is related to electrical behavior of heart.
F. J. Martfnez-Tabares D. H. Peluffo-Ord6nez C. Castro-Hoyos G. Castellanos-Dominguez
Universidad Nacional de Colombia sede Manizales, Caldas, Colombia
国际会议
上海
英文
774-777
2011-10-15（万方平台首次上网日期，不代表论文的发表时间）