An Optimized Non-Invasive Approach for Fetal Electrocardiogram Extraction using Wavelet Transformation and Thresholding

The importance of fetal electrocardiogram (FECG)extraction in prenatal care has increased to assess the health of the fetus and to monitor its growth. It helps in identifying any abnormalities and it can be cured or treated appropriately to ensure safe birth of the child. Over the years, a lot of methods have been developed to serve the purpose. Invasive methods are often risky because of the direct contact the electrodes make with the sensitive fetal scalp. Hence, research of non-invasive methods has been conducted widespread which resulted in the development of many new hybrid techniques. In this paper, a model was developed using the Continuous Wavelet Transform(CWT)to extract FECG from the abdominal electrocardiogram (AECG) by setting up appropriate scale value. In this method, continuous wavelet transformation is applied to get the fetal ECG and fetal Rpeaks were identified using wavelet thresholding. The extraction method was simulated in Python software. The dataset was imported from the online database, Physio Bank. The output has been validated using appropriate evaluation metrics and the built model proved consistent and reliable throughout ten different real-life datasets with an average accuracy of 91% and an F1 score of 95.7%.