Development of fetal electrocardiogram extraction based on blind source separation
Loading...
Date
2019-06
Authors
Fatin Izzati Binti Mohamad Abdul Hadi
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Electrocardiogram (ECG) test are commonly taken by pregnant women
to check the medical condition of the fetus during the pregnancy period in order to
prevent the occurrence of any cases of mortality and neonatal death. Fetal
Electrocardiogram (FECG) is a one dimensional signal used to monitor cardiac activity
and fetal. In order to collect ECG signal, electrodes are placed on thorax and abdomen
on mother‟s body. Due to the amplitude of FECG, it is small and easy to contaminate
with maternal Electrocardiogram (MECG), hence FECG must be extracted from the
mixed MECG and FECG signal before the analysis is carried out. Blind Source
Separation (BSS) is designed to extract desired signal without knowing any source
signal information. In this study, FastICA with Kurtosis and FastICA with negentropy
were used for FECG extraction. Performance evaluation is described in subjective
evaluation based on estimated signals characteristic and objective evaluation based on
signal to interference ratio (SIR) measurement parameter. By using five sets ECG data
taken from PhysioNet online website, the average value for five measurements is
calculated. From the experiment result, based on SIR measurement, FastICA with
negentropy shows better result than FastICA with Kurtosis. So, FastICA with
negentropy is recommended in FECG extraction. For Guide User Interface, the FECG
and MECG Monitoring Application was developed in order to give benefit for pregnant
women to monitor their fetal heart rate condition at home or anywhere.