Title: Machine learning based prediction model of the prolonged second stage of labor and its clinical validation
Background: The presence of the prolonged second stage of labor requires using instrumental delivery or transferring to cesarean delivery, which could have some adverse effects on maternal and newborn outcomes. It is important to identify risk factors for the prolonged second stage of labor and then to improve the success rate of vaginal delivery. The purpose of this study is to use machine learning algorithms to develop a model for predicting the presence of the prolonged second stage of labor.
Methods: This is a retrospective case-control study. We will extract the data of women from the electronic medical record system of a tertiary women’s and child’s hospital in China, including data of sociodemographic information and various perinatal information. The case group will be women who had suffered the prolonged second stage of labor in the past three years, while the control group will be women who delivered vaginally successfully. Five commonly used machine learning algorithms (including neural network, support vector machine, logical regression, decision tree, random forest) will be performed to construct the prediction model. We will randomly select 80% of the samples to construct the model, and the remaining 20% of the samples will be used to validate the prediction models of the prolonged second stage of labor and select the best model.
Discussion? This study will develop and validate the prediction model of the prolonged second stage of labor. The final model will provide a tool for predicting the prolonged second stage of labor in clinical practice and provide evidence for designing interventions to improve maternal and newborn outcomes.
1. This project is based on five machine learning algorithms, including a variety of clinical data and socio-demographic data of pregnant women, and plans to build a risk prediction model for the prolonged second stage of labor.
2. This study will investigate the incidence of the prolonged second stage of labor and its influencing factors, and analyze the impact of prolonged second stage of labor on maternal and infant outcomes.
This study will provide a tool for clinical prediction of prolonged risk in the second stage of labor, and also provide ideas for designing intervention measures, so as to promote maternal and fetal safety and reduce cesarean section rate.