Abstract

NARVAEZ-SANCHEZ, Raúl  and  JARAMILLO VELASQUEZ, Andrés. Differentiation between normal and arythmic electrocardiograms using frequency analysis. Rev. Cienc. Salud [online]. 2004, vol.2, n.2, pp.139-155. ISSN 1692-7273.

We report an advance of a research project based on the hypothesis that electrocardiograms (ECG) are easier to interpret in the frequency domain than in the time domain. This paper is aimed to medical personnel. Methodology: 21 electrocardiograms (eighteen normal and three arrhythmic) were taken from a database of Massachusetts Institute of Technology, and analyzed in the frequency domain, in order to distinguish quantitative and qualitatively spectra of arrhythmic patients from spectra of healthy patients. We worked on a small number of samples looking for stationary conditions, calculating the power spectral density (PSD) on intervals of 300 samples, which enclose two or three cardiac cycles, over a total range of 90.000 samples. For qualitative identification, the correspondent drafts were etched and their morphology compared. For quantitative identification, an average of the PSD of healthy registries we made, to which a distance were measured (how much an arrhythmic signal differs from the average of healthy signals) and differences of its integral to each arrhythmic PSD. Results: We found, preliminarily, differences between healthy and non healthy signals. The paper presents the drafts, distances and integrals of PSD of studied registries. Conclusion: In the frequency domain is possible to establish morphologic and numeric differences, easy to identify, between healthy and non healthy electrocardiograms.

Keywords : Physiology; electrocardiogram; signal processing; computer-assisted.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )