A novel algorithm for early detection of fetal arrhythmia using ICA

Fetal ECG monitoring provides important information about the condition of the fetus and it helps during labour. In extreme cases, a severe fetal arrhythmia can cause collapse or sudden cardiac death. During pregnancy period fetal health must be carefully monitored to detect early fetal cardiac diseases, and for providing treatment as soon as possible. At present reference based approaches are used to extract fetal ECG from the mother's abdomen ECG. But this ECG contains artifacts due to the muscle contraction, breathing, and noises. This makes even harder to extract the ECG. The research going to extract fetal ECG using the method BSS. In this paper proposes a novel algorithm for early detection of fetal Electrocardiogram (ECG) analysis and early detection of fetal arrhythmia. The Independent component analysis (ICA) algorithm is used to preprocess and extract the fetal ECG. Peak detection algorithm using state-machine logic is used to extract the features from the fetal ECG. The classification is done by using Naïve Bayesian classifier. Most of the data used for this study were extracted from the Physionet database. Validation of the proposed extraction and detection framework has been done on publicly available datasets, showing promising results (Specificity = 96.29%, accuracy = 93.71%).