Postpartum hemorrhage risk prediction method and early warning system based on fetal monitoring uterine contraction graph and high-risk factors

Abstract

The invention discloses a postpartum hemorrhage risk prediction method and early warning system based on a fetal supervision uterine contraction graph and high risk factors. The method comprises the following steps: S1, collecting data of a hospital fetal supervision uterine contraction graph and an electronic medical record system; S2, screening pregnant and lying-in women during vaginal delivery, and extracting required patient information; S3, establishing a hemorrhage risk prediction model within 24 hours after delivery of the puerpera by applying a machine learning method; S4, evaluating the prediction value of each model, and selecting an optimal prediction model. According to the method, the prediction model is established through a machine learning method, and the method is of great significance to clinical treatment decisions including timely identification of uterine contraction hypodynamia, prevention of soft birth canal injury, correction of blood coagulation dysfunction and the like and reduction of the occurrence rate of bleeding.

Description

technical field [0001] The invention relates to the technical field of obstetrical clinical risk prediction, in particular to a postpartum hemorrhage risk prediction method and an early warning system based on fetal monitoring uterine contractions and high-risk factors. Background technique [0002] With the development of medical information electronic technology, the clinical important indicators of pregnant women and the information of previous pregnancy tests are all recorded in electronic medical records, making it possible to extract some high-risk factors and data of postpartum hemorrhage, which provides doctors with a comprehensive grasp It is convenient to predict the risk of postpartum hemorrhage according to the physical condition of pregnant women. However, due to the numerous and complex factors related to postpartum hemorrhage, it is difficult for even experienced doctors to comprehensively and objectively analyze these large clinical data with heterogeneous ty...

AI Technical Summary

Problems solved by technology

[0004] In the prior art, traditional statistical methods (logistic regression) are mostly used to carry out postpartum hemorrhage prediction modeling research on a specific group of people, but no data mining and machine learning methods with stronger fitting capabilities are used for modeling research. At the same time, there is no relevant postpartum hemorrhage prediction research that includes uterine contractography (Tocography, TOCO) as the modeling object

Images