The Research Of Fetal ECG Detection Method And Fetal Monitoring System

Fetal ECG monitoring is one of the most effective methods of fetal monitoring, which is to monitor fetal ECG during pregnancy and analysis fetal heart activity. Fetal monitoring is an important way to ensure pregnant women and fetus safety and achieve prenatal and postnatal care now. It consists of two parts:the extraction of fetal ECG and the development of fetal ECG monitoring systems. They are both the current research focus.This research work was mainly directed against the two parts above. The problem of fetal ECG extraction was raised, mainly due to fetal ECG in maternal abdominal signal acquired by non-invasive electrode method containing much undesired signal. Especially maternal ECG brought great difficulties for fetal ECG extraction. Based on the complex characteristics of maternal abdominal signal, extracting fetal ECG is very important not only in clinical diagnosis, but also in signal processing field, which is a typical problem of signal processing and detection. In this paper the adaptive filtering and wavelet transform method was discussed to solve this complex issue. Another part of this research was the construction of fetal ECG monitoring system. A PC-based fetal ECG monitoring system was developed in this work. The system had data acquisition and real-time processing capabilities, which was real-time fetal ECG extraction on the basis of the algorithm. In addition, a case library was needed in the system to achieve user information management.Adaptive filtering is the most commonly used method to extract fetal ECG now. The method required two channels of signal respectively acquired from maternal abdomen and chest, the abdominal signal was original input signal, the pectoral signal was reference input signal. Through the adaptive process, adjusting the coefficients of adaptive filter constantly, the system could get the best output, and fetal ECG was extracted. Adaptive filtering algorithm has the characteristic of less computation, is simple and suitable for the requirements of real-time processing.Wavelet transform is a more popular method of signal processing now. Wavelet transform modulus maximum method was used to extract fetal ECG from maternal abdominal signal in this paper. The method could identify fetal ECG from only maternal abdominal signal, which needn't maternal pectoral signal as reference signal. The algorithm has large amount of computation and defect in terms of real-time. A lot of work had been done to optimize the algorithm, but it was still hard to get satisfactory results. Further improvement of the algorithm is needed.A PC-based fetal ECG monitoring system was composed of hardware and software two parts. Hardware part was constructed by the ECG amplifier and AD acquisition card, to achieve signal amplification and A/D conversion; software parts adopted the Windows XP system, C# and SQL Server to program, to control data acquisition of AD card, process data and manage users'information.