Key result
A stacking-based machine learning model predicted 30-day all-cause hospital readmissions in patients with acute myocardial infarction with an AUC of 0.720, outperforming the best individual baseline model (XGBoost, AUC 0.713).
Why the study?
Acute myocardial infarction is followed by a high 30-day readmission rate, and accurate prediction is needed to identify high-risk patients and optimize medical resources.
Does a stacking-based machine learning model improve the prediction of 30-day all-cause hospital readmissions in patients with acute myocardial infarction compared to individual machine learning models?
Observational (n=3,283)
No
Does a stacking-based machine learning model improve the prediction of 30-day all-cause hospital readmissions in patients with acute myocardial infarction compared to individual machine learning models?
Absolute Event Rate: 0.72% vs 0.713%
A stacking-based machine learning model using clinical data can effectively predict the risk of 30-day all-cause hospital readmissions for patients with acute myocardial infarction, outperforming individual machine learning classifiers.
May aid 30-day AMI readmission risk stratification in resource-limited settings; leaves open need for external validation and outcome trials.
BACKGROUND: Acute myocardial infarction (AMI) is a serious cardiovascular disease, followed by a high readmission rate within 30-days of discharge. Accurate prediction of AMI readmission is a crucial way to identify the high-risk group and optimize the distribution of medical resources. METHODS: In this study, we propose a stacking-based model to predict the risk of 30-day unplanned all-cause hospital readmissions for AMI patients based on clinical data. Firstly, we conducted an under-sampling method of neighborhood cleaning rule (NCR) to alleviate the class imbalance and then utilized a feature selection method of SelectFromModel (SFM) to select effective features. Secondly, we adopted a self-adaptive approach to select base classifiers from eight candidate models according to their performances in datasets. Finally, we constructed a three-layer stacking model in which layer 1 and layer 2 were base-layer and level 3 was meta-layer. The predictions of the base-layer were used to train the meta-layer in order to make the final forecast. RESULTS: The results show that the proposed model exhibits the highest AUC (0.720), which is higher than that of decision tree (0.681), support vector machine (0.707), random forest (0.701), extra trees (0.709), adaBoost (0.702), bootstrap aggregating (0.704), gradient boosting decision tree (0.710) and extreme gradient enhancement (0.713). CONCLUSION: It is evident that our model could effectively predict the risk of 30-day all cause hospital readmissions for AMI patients and provide decision support for the administration.
No takes yet. Share an insight, caveat, or question.
Zhang et al. (2020) conducted an observational in Acute myocardial infarction (AMI) (n=3,283). Stacking-based machine learning model vs. Individual machine learning models (e.g., XGBoost, Random Forest, SVM, Decision Tree) was evaluated on Area under the receiver operating characteristic curve (AUC) for predicting 30-day all-cause hospital readmission. A stacking-based machine learning model predicted 30-day all-cause hospital readmissions in patients with acute myocardial infarction with an AUC of 0.720, outperforming the best individual baseline model (XGBoost, AUC 0.713).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: