Fetal ECG extraction and enhancement in prenatal monitoring — Review and implementation issues

Abstract

This paper presents a detail review and implementation issue of fetal ECG extraction and enhancement. The focus is also made on proper placement of electrodes for fetal ECG monitoring in twins and multi-fetal prenatal. Various extraction methods like correlation, subtraction, matched filtering, linear regression and independent component analysis are discussed. For enhancement neural networks, fuzzy logic, wavelet transform and polynomial networks are discussed. When methods like correlation and subtraction are obsolete, adaptive filtering and independent component analysis methods are mainly used for extraction and enhancement of FECG (Fetal electro cardiogram). The enhancement methods are used for classification and diagnosis purpose through parameter analysis. The method to be used depends on the number of electrodes used. If more number of electrodes are used the accuracy is more but the complexity increases. To overcome the tradeoffs between complexity and accuracy a new method is proposed to minimize the complexity in placing the electrodes and a modified independent component analysis method is proposed for extraction. For enhancement a new algorithm is proposed to improve the quality of Fetal ECG using signal processing technique.