Research On Fetal Signal Processing And Its Implementation In Android Mobile Terminal

Fetal movement in fetal physiological indicators is one of the most intuitive form of expression.Fetal exercise can be used to monitor the direct health of the fetus and to understand its neurodevelopmental status.There are two main ways to detect fetal movement: the hospital ultrasound and pregnant women self-perception.Hospital ultrasound testing equipment is expensive,but if the long-term ultrasound testing may have a certain impact on the fetus,while the mother perceived fetal movement can only feel 36% of fetal movement detected by ultrasound imaging.In view of the clinical significance of fetal movement reduction,there are a large number of scholars at home and abroad have conducted many studies on how to identify fetal movement.So far,the practical domestic fetal motion monitoring equipment in the market is still blank.With the new acceleration sensor technology is mature,it has been in the medical detection direction has been well developed.This topic will study the use of specially manufactured suture with acceleration sensor to collect fetus movement signal,through the micro-control processor,the original data of fetal movement will be transmitted to Android mobile terminal equipment through Bluetooth low energy analysis and identification,and finally the number of fetal movement visualization.But the fetal signal contains a lot of false information caused by the maternal action or other human factors that may reduce the signal quality,thus causing a great obstacle to the detection of fetal movement.To this end,this paper has done the following work to realize low cost,portability and intelligent mobile fetal motion detection and monitoring equipment:1)In this paper,four acceleration sensors are used to collect the fetal movement signals on the abdominal wall of pregnant women.The data is packaged and processed by the micro control processor and then controlled by Bluetooth to send the data to the Android mobile device.2)In this paper,the commonly used methods of signal time-frequency analysis are analyzed and analyzed.The method of introducing improved B distribution for the crosstalk term of Wigner-Ville distribution is introduced.The characteristics of fetal signal and mother jamming signal are also analyzed.The characteristics of fetal signal and interference signal are extracted,and the FIR band-pass filter is designed according to the characteristics of fetal signal.The filtered signal is further processed by wavelet denoising.Finally,the filtered and denoised signal is sent to BP neural network for learning and training,and finally fetal movement signal is identified.3)This article in the Android Studio development platform to achieve a Bluetooth mobile communications terminal and fetal signal acquisition platform,and the received fetal movement data stored in the phone memory.Through the NDK BP neural network algorithm is ported to the Android system,and ultimately can be stored in the phone fetal movement data for fetal movement identification,and fetal movement times displayed on the mobile client.