[[getLangText('background')]]
Preterm birth (PTB), defined as delivery before 37 weeks of gestation, affects approximately 15 million infants globally each year, accounting for about 11% of all live births. It is a leading cause of neonatal mortality and long-term morbidity, with about 35% of neonatal deaths attributed to PTB. In China, the incidence of PTB is around 7.8%, resulting in significant health challenges and economic burdens. While advanced maternal age is a known risk factor, young women under 35 also face considerable risk due to factors such as teenage pregnancy, malnutrition, and lower socioeconomic status. Early and accurate prediction of PTB risk is crucial for timely intervention and improved maternal and neonatal health outcomes.
[[getLangText('parameter_description')]]
ALP:An enzyme related to liver function, important for PTB risk assessment.
Na:Essential electrolyte balance, influencing maternal and fetal health.
AFP:A protein produced by the fetus, with elevated levels indicating potential complications.
UREA:Indicates kidney function and metabolic state, critical for assessing overall maternal health.
Lym#:Absolute value of lymphocytes, crucial for evaluating the immune response.
HGB:A measure of blood's oxygen-carrying capacity, important for detecting anemia-related risks.
RDWCV: A measure of the variation in red blood cell size, linked to various health conditions.
The Under-35 Preterm Birth Predictor (U35-PBP) utilizes advanced machine learning algorithms to provide real-time risk assessments for preterm birth in pregnant women under 35. Based on a comprehensive analysis of clinical and laboratory data, the U35-PBP model incorporates key predictive variables identified through rigorous statistical and machine learning techniques. The model was trained and validated using a robust dataset from Hangzhou First People's Hospital, demonstrating high accuracy and reliability.
Principle:
First divide the data into training set and test set, then use the cross-validation method to train the model in the training set, train the optimal model as the final model and record the threshold at this time as the final threshold, and finally observe the model in the test performance on the set. By continuously adjusting the parameters of the model, the generalization ability of the model is improved, and the performance of the model in the training set, validation set and test set is relatively optimal.
By the predicted sample into the optimal model, the model will predict the probability of occurrence, generate the SHAP force plot, and then evaluate the risk according to the predicted probability and prediction.