Open AccessDETECTION OF VENTRICULAR FIBRILLATION USING WAVELET TRANSFORM AND PHASE SPACE RECONSTRUCTION FROM ECG SIGNALS
Author(s) -
Seok-Woo Jang,
Sang-Hong Lee
Publication year - 2021
Publication title -
journal of mechanics in medicine and biology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.236
H-Index - 30
eISSN - 1793-6810
pISSN - 0219-5194
DOI - 10.1142/s0219519421400364
Subject(s) - pattern recognition (psychology) , wavelet , normal sinus rhythm , artificial intelligence , ventricular fibrillation , computer science , wavelet transform , noise (video) , fuzzy logic , mathematics , medicine , cardiology , atrial fibrillation , image (mathematics)
This study proposes the detection of ventricular fibrillation (VF) by wavelet transforms (WTs) and phase space reconstruction (PSR) from electrocardiogram (ECG) signals. A neural network with weighted fuzzy memberships (NEWFM) is used to detect VF as a classifier. In the first step, the WT was used to remove noise in ECG signals. In the second step, coordinates were mapped from the wavelet coefficients by the PSR. In the final step, NEWFM used the mapped coordinates-based features as inputs. The NEWFM has the bounded sum of weighted fuzzy memberships (BSWFM) that can easily appear the distinctness between the normal sinus rhythm (NSR) and VF in the graphical characteristics. The BSWFM can easily be set up in a portable automatic external defibrillator (AED) to detect VF in an emergency.