RESUMO
Background: This present work focused on predicting prognostic outcome of inpatients developing acute exacerbation of chronic obstructive pulmonary disease (AECOPD), and enhancing patient monitoring and treatment by using objective clinical indicators. Methods: The present retrospective study enrolled 322 AECOPD patients. Registry data downloaded based on COPD Pay-for-Performance Program database from January 2012 to December 2018 were used to check whether the enrolled patients were eligible. Our primary and secondary outcomes were ICU admission and in-hospital mortality, respectively. The best feature subset was chosen by recursive feature elimination. Moreover, seven machine learning (ML) models were trained for forecasting ICU admission among AECOPD patients, and the model with the most excellent performance was used. Results: According to our findings, random forest (RF) model showed superb discrimination performance, and the values of area under curve (AUC) were 0.973 and 0.828 in training and test cohorts, separately. Additionally, according to decision curve analysis, the net benefit of RF model was higher when differentiating patients with a high risk of ICU admission at a <0.55 threshold probability. Moreover, the ML-based prediction model was also constructed to predict in-hospital mortality, and it showed excellent calibration and discrimination capacities. Conclusion: The ML model was highly accurate in assessing the ICU admission and in-hospital mortality risk for AECOPD cases. Maintenance of model interpretability helped effectively provide accurate and lucid risk prediction of different individuals.
