Title :
Fetal QRS extraction based on wavelet analysis and pattern matching
Author :
Echeverria, J.C. ; Ramírez, Norberto ; Pimentel, Ana B. ; Rodríguez, Richardo ; Gonzalez, Ramon ; Medina, Veronica
Author_Institution :
Dept. de Ingenieria Electr., Univ. Autonoma Metropolitana, Mexico City, Mexico
fDate :
31 Oct-3 Nov 1996
Abstract :
A high precision method for fetal QRS complex extraction from the abdominal ECG is presented. The method is based on a wavelet multiresolution decomposition and a pattern matching procedure. The wavelet analysis allowed the elimination of detail signals that did not have maternal and fetal QRS frequency components and it also allowed maternal and fetal complexes homogenization. The existence and position of a fetal QRS immersed in a maternal QRS was determined by correlating the maternal complex with a set of cases that represent all possible fetal occurrences. The method was able to detect all immersed fetal QRS complexes in 5 real signals. Non significant differences for the position of 202 extracted fetal complexes were found on simulated signals. This method introduces a new approach resulting in excellent fetal QRS extraction and maternal QRS cancellation
Keywords :
electrocardiography; feature extraction; medical signal processing; pattern matching; wavelet transforms; abdominal ECG; electrodiagnostics; fetal QRS extraction; fetal QRS frequency components; maternal QRS frequency components; maternal/fetal complexes homogenization; pattern matching; wavelet analysis; wavelet multiresolution decomposition; Abdomen; Electrocardiography; Fetal heart rate; Frequency; Pattern analysis; Pattern matching; Signal processing; Signal resolution; Signal to noise ratio; Wavelet analysis;
Conference_Titel :
Engineering in Medicine and Biology Society, 1996. Bridging Disciplines for Biomedicine. Proceedings of the 18th Annual International Conference of the IEEE
Conference_Location :
Amsterdam
Print_ISBN :
0-7803-3811-1
DOI :
10.1109/IEMBS.1996.647598