A design of fetal monitoring system based on the time-frequency analysis of fetal heart beat fluctuation using wavelet transform

我们研究了胎儿自主神经系统的非线性控制机制是否会在不同的胎儿状态下发生变化。正常胎、低氧胎和胎儿中毒胎均检测到8000次以上胎心跳动。采用高精度自相关法测定胎儿心脏多普勒信号间隔，得到胎儿心率波动的时间序列。采用时频分析方法小波变换研究了胎儿心率频率低频分量中时间波动幅度的分布。样条4作为母小波函数。从妊娠17周开始观察到伽玛分布。该gamma分布的参数upsilon的值约为1.6，并且与妊娠年龄或一天中的时间无关，其值保持不变。胎儿发生酸血症时，upsilon值显著下降至0.77，低氧血症时为1.51，正常情况下为1.54。本研究阐明了胎儿心率波动时间序列的非线性结构。该技术可为胎儿酸血症的诊断提供一种新的胎儿监测定量指标。

We examined whether the non-linear control mechanism of the fetal autonomic nervous system would change in various fetal states. 8000 or more fetal heart beats were detected from normal, hypoxemic and acidemic fetuses. Fetal heart Doppler signal intervals were determined in a high-precision autocorrelation method and a time series of fetal heart rate fluctuation was obtained. The distribution of the amplitude of temporal fluctuation in the low frequency component of fetal heart rate frequency was studied using a method of time-frequency analysis called wavelet transform. Spline 4 was used as the mother wavelet function. A gamma distribution was observed from 17 weeks of gestation onward. The value of the parameter upsilon of this gamma distribution was approximately 1.6 and remained constant regardless of the gestation age or the time of the day. The value of upsilon decreased significantly to 0.77 when the fetus developed acidemia, and was 1.51 in hypoxemia and 1.54 in a normal condition.This study elucidates a non-linear structure of the time series of heart rate fluctuation of the gamma distribution in human fetus. This technique may provide a new quantitative index of fetal monitoring to diagnose fetal acidemia.