.png){kind=icon}
.png){kind=icon}
Return
Medication-based mortality prediction in COPD using machine learning and conventional statistical methods
.png){kind=icon}
OA
AI
DOI：10.1016/j.ijmedinf.2025.106177.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Prediction tools often need data not available electronically during routine visits. • Medication claims enable real-time risk stratification during routine visits. • Deep neural networks slightly outperformed logistic regression models. • Including comorbidity medications improved model performance significantly. • SHAP highlighted non-inhaled anticholinergics and diuretics among top predictors.
Keywords:
COPD management
Medication adherence
Machine learning
Deep neural network
Prediction model
AUC-ROC
Area under the receiver operating characteristic curve
AUC-PR
Area under the precision-recall curve
CE
Cohort entry
CI
Confidence interval
COPD
Chronic obstructive pulmonary disease
D-ANN
Deep artificial neural network
ICS
Inhaled corticosteroid
LAAC
Long-acting anticholinergic
LABA
Long-acting β2-agonist
LASSO
Least absolute shrinkage and selection operator
LTRA
Leukotriene-receptor antagonist
METH
Methylxanthines
ML
Machine learning
OCS
Oral corticosteroid
PDC
Proportion of days covered
RA
Respiratory antibiotics
RAMQ
Régie de l’assurance maladie du Québec
RF
Random forest
SAAC
Short-acting anticholinergic
SABA
Short-acting β2-agonist
SHAP
SHapley Additive exPlanations
S-ANN
Simple artificial neural network
XGBoost
Extreme Gradient Boosting
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.1
Papers:
4.5K
Citations:
1.1W