Research on Giant Child Risk Prediction Based on Machine Learning Models

This study collected clinical data from 3233 Han Chinese singleton pregnant women who were admitted to the hospital between April 2024 and March 2025. The training and testing sets were divided in an 8:2 ratio, and samples were randomly selected as an independent external validation set. The study used five machine learning algorithms, including Multi Layer Perceptron (MLP) and Random Forest, to establish classification models. The performance of the models was systematically evaluated based on indicators such as Area Under the Receiver Operating Characteristic Curve (AUC), sensitivity, and specificity. To further enhance clinical interpretability, the SHapley additive interpretation (SHAP) method is used to quantify the contribution of each feature to the model output, revealing the key biological driving factors of macrosomia. This methodological framework integrates complex data modeling with clinical decision support, providing operational intelligent risk assessment tools for primary hospitals.