Nonlinear processing and analysis of ECG data

Article type: Research Article

Authors: Tu, Chengyuan^a | Zeng, Yanjun^{a; *} | Yang, Xinchun^b

Affiliations: [a] Biomechanics & Medical Information Institute, Beijing Polytechnic University, P.R. China 100022 | [b] Department of Internal Medicine, Beijing Chao-Yang Hospital, P.R. China 100026

Correspondence: [*] Corresponding author: E-mail: [email protected].

Abstract: A new approach based on a histogram analysis to detect QRS-T complexes of the ECG curve rapidly and reliably is described. The aim consists of a rapid and reliable detection of the P-wave (when atrial fibrillation, AF, is absent) or the f-wave (when AF happens). By means of signal processing techniques such as the fractal dimension method and the Shannon entropy method, two types of ECG signals (with or without AF) were analysed and the significant differences between them documented. Accordingly, the methods allow to determine easily whether AF occurs or not.

Keywords: ECG curve, P-wave, f-wave, AF (atrial fibrillation), histogram, fractral dimension, Shannon entropy

DOI: 10.3233/THC-2004-12101

Journal: Technology and Health Care, vol. 12, no. 1, pp. 1-9, 2004