Fetal Heart Rate Extraction And System Design Method Based On EMD And Wavelet Time-frequency Analysis

Prenatal and postnatal care and healthy Chinese strategies are the topics of livelihood that the government has been paying more attention to.The health of newborns is related to the well-being of each family and it is related to the overall development of a new generation of young people in the future.Fetal heart rate monitoring can ensure the health of newborns to a certain extent.For example,the fetal heart rate value can reflect whether the fetus is in the mother’s womb whether there is oxygen deficiency or other information.According to this index,it can quickly determine whether the fetus is in a bad condition.Therefore,the extraction of fetal heart rate is a very significant part of the fetal heart rate monitoring process.In this context,the paper proposes a fetal heart rate extraction method based on EMD and wavelet time-frequency analysis,and also designs a fetal heart signal acquisition system based on the laboratory platform.The research on the fetal heart signals found that the Doppler fetal heart signal is a non-stationary and non-linear random signal,which is influenced by the environment of the mother’s womb and is a signal with low signal-to-noise ratio.In the existing research methods,the wavelet transform has good effect on the fetal heart signal preprocessing,but there are some factors such as interference due to the uncertainty of the wavelet basis,and the preprocessing method of the independent component analysis is difficult to separate effective fetal heart signals directly in the case of low signal to noise ratio.Therefore,this paper proposes a new fetal heart rate extraction method based on this situation.The main research work and innovations of this paper are as follows:1.According to the existing problems in the pretreatment method of fetal heart signals,a new signal denoising method based on Empirical Mode Decomposition(EMD)and wavelet energy distribution is proposed.The paper is validated by multiple sets of experiments.The effectiveness of this algorithm in denoising the fetal heart signal was compared with the typical two pretreatment methods to verify its superiority.2.Since the cycle of pre-processed fetal heart signals are usually extracted through the time domain autocorrelation,the paper tries to analyze the fetal heart signals from the time-frequency analysis method by using the time-frequency analysis method of wavelet transform.Simulation experiments show that fetal heart rate can be successfully extracted in this method,and compared it with the time domain autocorrelation method.3.The paper finally designs a set of fetal heart signal acquisition system based on the laboratory platform.This system uses an ultrasonic detection scanning device based on program-controlled distributed motor drive,which can more conveniently and accurately locate the fetal heart.The principle of this system and its hardware structure are analyzed in detail.