Fetal heart rate extraction using normalized least mean square (nlms) algorithm
Loading...
Date
2019-06
Authors
Ricky Kiing Jiu Yuan
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The purpose of this project is to develop a fetal heart rate (FHR) extraction application to analyze the fetus activity in the mother uterus. There are several methods that can use to detect FHR. One example is using the fetal electrocardiogram (FECG) that generated by fetus’ heart. Extracting FECG signals while the fetus in the mother uterus is consider a major challenge. Before interpreting the condition of FHR, the input signal is required to compute the fetal heartbeat per minute bpm. At the beginning of the process, the input signal will pass through adaptive filter to get FECG. Normalized Least Mean Square (NLMS) algorithm is one of adaptive filters and is chosen for this thesis. After that, use Pan Tompkins algorithm technique to track R-peaks (heartbeat pulse) in FECG signal. Whenever the RR interval is detected, a formula is used to calculate the bpm of FECG. Abdominal and direct FECG (ADFECG) database will be used to evaluate the implemented techniques as it has reference signal. At the end of research, the FHR calculated is varied from 126 bpm to 130 bpm. When comparison is done between abdominal ECG (AECG) and direct FECG (DFECG), the percentage error of FHR is 0.1%. The accuracy of R-peaks extraction is 100% where all Rpeaks are detected by implemented techniques. All implemented techniques are used in a graphical user interface (GUI) in MATLAB. This system will have ability to interpret the non-invasive FECG (NIFECG) database and compute its FHR.