Extraction of Fetal Electrocardiographic Signals Using Neural Network

摘要

Importance of fetal monitoring is undisputed during pregnancy. Fetal electrocardiography (fECG) is preferred over Doppler ultrasound for its ease of use and continuous monitoring. However, utility of fECG is possible only when a reliable recording can be obtained. Currently used fECG modalities gives low amplitude signals and are corrupted by different sources of interferences like maternal ECG (mECG), power line interference, motion artifacts, base line wander and different types of noises. Therefore, appropriate signal processing techniques are required to reveal better fetal ECG from abdominal ECG. The present study was conducted to extract fECG from abdominal ECG (aECG) using neural network.