Meteorological factor-driven prediction of high-use days of budesonide: construction and comparison of ensemble learning models
- VernacularTitle:气象因素驱动的布地奈德高用药日预测集成学习模型的构建与比较
- Author:
Qitao CHEN
1
;
Yue ZHOU
2
;
Xiaojun ZHANG
2
;
Jingwen NI
2
;
Guoqiang SUN
2
;
Fenfei GAO
3
;
Lizhen XIA
2
;
Zihao LI
3
Author Information
1. Medical College，Shantou University，Guangdong Shantou 515041，China;Dept. of Pharmacy，Sanming Hospital of Integrated Traditional Chinese and Western Medicine，Fujian Sanming 365000，China
2. Dept. of Pharmacy，Sanming Hospital of Integrated Traditional Chinese and Western Medicine，Fujian Sanming 365000，China
3. Medical College，Shantou University，Guangdong Shantou 515041，China
- Publication Type:Journal Article
- Keywords:
budesonide;
meteorological factors;
ensemble
- From:
China Pharmacy
2025;36(21):2723-2726
- CountryChina
- Language:Chinese
-
Abstract:
OBJECTIVE To construct ensemble learning models for predicting high-use days of budesonide based on meteorological factors， thereby providing reference for hospital pharmacy management. METHODS Meteorological data for 2024 and outpatient budesonide usage data from the jurisdiction of Sanming Hospital of Integrated Traditional Chinese and Western Medicine were collected. High-use days were defined as the 75th percentile of outpatient budesonide usage， and a corresponding dataset was established. The prediction task was formulated as a classification problem， and three ensemble learning models were developed： Random Forest， Extreme Gradient Boosting （XGBoost）， and Histogram-based Gradient Boosting Classifier. Model performance was evaluated using accuracy， precision， recall， F1-score， and log-loss. Model interpretability was analyzed using Shapley Additive Explanations （SHAP）. RESULTS The Histogram-based Gradient Boosting Classifier achieved the best performance （accuracy＝0.75， F1-score＝0.48）， followed by XGBoost （accuracy＝0.74， F1-score＝0.43） and Random Forest （accuracy＝0.72， F1-score＝0.22）. SHAP results suggested that the prediction results of the last two models have the highest correction. CONCLUSIONS Ensemble learning models can effectively predict high-use days of budesonide， with the Histogram- based Gradient Boosting Classifier demonstrating the best predictive performance. Low temperature， high humidity， and low atmospheric pressure show significant positive impacts on the prediction of daily budesonide usage.
