The Design and Implementation of Intelligent Labor Contraction Monitoring System based on Wearable Internet of Things

Abstract

In current clinical practice, pregnant women who have entered 37 weeks cannot correctly judge whether they are in labor based on their subjective feelings. Wrong judgment of labor contraction can lead to adverse pregnancy outcomes and endanger the safety of mothers and babies. It will also increase the healthcare pressure in the hospital and the healthcare efficiency is reduced. Therefore, it is very meaningful to be able to design a system for monitoring labor contraction based on objective data to assist pregnant women who have entered 37 weeks in deciding the suitable time to go to hospital. For the above requirements, this thesis designs and implements an intelligent labor contraction monitoring system based on wearable Internet of Things. The system combines the Internet of Things technology, wearable technology and machine learning technology to collect contraction data through wearable sensing device. It uses the Long Short-Term Memory (LSTM) neural network to classify and identify the collected contraction data and realize real-time processing. It improves the accuracy of model recognition to 93.75%. And the recognition results are fed back to the WeChat applet so that pregnant women can view them in real time. The prototype of the wearable sensing device has been integrated by 3D printing and the proof-of-concept system has been demonstrated. Pregnant women can use this system to detect the contraction status and view the contractions in real time through the WeChat applet results. They can judge whether it is suitable for labor, and this system assists in making decisions about the best time to go to hospital.