| 1.IntroductionFetal Heart Rate signal is an objective index which reflects the physiologicalactivity of foetus heart electricity, It signifies an important life parameter that thelife entity exists or not, it is a kind of typical biomedical signal, belong to the non-steady random signal. Now, For dealing with the signal, There are some ways, forexample Wiener filters, concerned with average, space flold average, correlation ,and so on.But these way can not get over efficiencyly the overturn of fetal heartrate curve.this paper, we adopt two microchip computer to gather and control thedoppler ultrasound fetal heart rate signal. And ensure the real time and veracity offetal heart data;the authors put forward an approach to extract fetal heart signalbased on coif5 wavelet transform and double thresholds. It's proved by theexperiment that this method improve the veracity of extracting fetal heart andovercome the overturn problem of the curve availably because of the FHR doubleand halve.2. The hardware system of design of extract heartbeat sound fromDoppler signal One most important parameters of fetal monitoring is the FHR (Fetal HeartRate) in the perinatal management. The ultrasound signal of Doppler is a goodnon-invasive way to detect the FHR, so in this paper ,we select the transducer ofDoppler . But it's difficult to extract the fetal heart signal because of thecomplexity of Doppler fetal signal and the system of the FHR collect the a greatdeal of the heartbeat sound data and the system of the FHR demand the high thereal-time. In this paper , the fetal heart signal was got by the circuit of dealingwith the signal . The sampling microchip computer control the MAX187 toA/D sample the heartbeat sound , and storage the data to the two portIDT7134.and the signal data come into the computer system by thecommunicating microchip computer .The computer system achieve thecommunicated controlling, the arithmetic of abstracting the hear beatsound .show the fetal heart rate monitoring or show warn function of abnormalfetal heart rate, the system has shows its high precision and real-time.when the system pretreatment the heartbeat sound from Doppler signal ,itprocess mostly lowpass ,absolute value operation and abstract the bag signal. Thelowpass adopt 250Hz the two steps lowpass , we use it to get rid of the highfrequency and noise. Because of The absolute value operation , the intensity ofthe hear beat sound signal become the double intensity, and advance theprecision of the measure. The circuit of abstracting the signal adopt the lowpassof the conjoint 10 Hz' П filter and T filter .The heartbeat sound signal istransformed the electricity-level signal by the electricity-level Compare and thesimulate level is transformed the high-low-level signal. These high-low-levelsignal only need the a bit by the microchip computer sampling, dispersing andfiguring. so this way may save the memory space of the microchip computer.When the fetal heart signal is adopted , storage and transfer, this systemelect the high speed and high precision MAX187 A/D of MAXIM company andthe fetal heart data is saved in high speed double port RAM IDT7134;and adoptthe communicated microchip computer transfer the fetal heart data to thecomputer, so the real-time of the system is advanced by the adopted andcommunicated of the system advancing at same time.Finally, the arithmetic of coif5 wavelet transform and double thresholdswas accomplished on the system of computer.3. Extract fetal heart signal based on coif5 wavelet transform anddouble thresholdsWavelet analysis is a rapid developed new area of nowadays appliedmathematics. Wavelet transform(WT) is space(time) and frequency local transformcompared to Fourier transform and Windowed Fourier transform(Gabor transform).So it can pick-up local information from signal and can realize multi-analysis tosignal by stretching , shrinking and shift etc. It solved many problems whichcan't be solved by Fourier transform. So we call wavelet transform as"mathematics microscope". Now wavelet transform is used widely in many areas.Especially in signal processing area it shows great advantage and has promisingfuture.The keystone of this paper is using the arithmetic of coif5 wavelet transformand double thresholds to abstract the fetal heart rate. To introduce the problem ofarithmetic of correlation and bring the overturn problem of the fetal heart rate curve.This paper analyze the superiority of wavelet dealing with non-steady randomsignal .Through analyzing the index of wavelet base function and comparing thecharacteristic of the wavelet base function, this paper select the coif5 wavelet basefunction abstracting the fetal heart signal;Through analyzing the frequencycharacteristic of the fetal heart signal and the trial effect of abstracting the fetalheart signal, the paper make sure the layer number of wavelet decompose. Whencounting the fetal heart rate, this paper put forward the theory of double thresholds.so from the theory, the paper put forward the superiority of Extracting fetal heartsignal based on coif5 wavelet transform and double thresholds.4. ConclusionOn computer simulation experiment , through abstracting the data of fetalheart sound, the author materialize the superiority of coif5 wavelet transform anddouble thresholds . It's proved by the experiment that this method improve theveracity of extracting fetal heart and overcome the overturn problem of the curveavailably because of the FHR double and halve. And Through competing thetraditional autocorrelation arithmetic and coif5 wavelet arithmetic, from thepractice apply, the author proved this arithmetic validity and superiority... |
PDF Full Text Request