Development and validation of a machine learning-based readmission risk prediction model for non-ST elevation myocardial infarction patients after percutaneous coronary intervention

Abstract

To investigate the factors that influence readmissions in patients with acute non-ST elevation myocardial infarction (NSTEMI) after percutaneous coronary intervention (PCI) by using multiple machine learning (ML) methods to establish a predictive model. In this study, 1576 NSTEMI patients who were hospitalized at the Affiliated Hospital of North Sichuan Medical College were selected as the research subjects. They were divided into two groups: the readmitted group and the non-readmitted group. The division was based on whether the patients experienced complications or another incident of myocardial infarction within one year after undergoing PCI. Common variables selected by univariate and multivariate logistic regression, LASSO regression, and random forest were used as independent influencing factors for NSTEMI patients' readmissions after PCI. Six different ML models were constructed using these common variables. The area under the ROC curve, accuracy, sensitivity, and specificity were used to evaluate the performance of the six ML models. Finally, the optimal model was selected, and a nomogram was created to visually represent its clinical effectiveness. Three different methods were used to select seven representative common variables. These variables were then utilized to construct six different ML models, which were subsequently compared. The findings indicated that the LR model exhibited the most optimal performance in terms of AUC, accuracy, sensitivity, and specificity. The outcome, admission mode (walking and non-walking), communication ability, CRP, TC, HDL, and LDL were identified as independent predicators of readmissions in NSTEMI patients after PCI. The prediction model constructed by the LR algorithm was the best. The established column graph model established proved to be effective in identifying high-risk groups with high accuracy and differentiation. It holds a specific predictive value for the occurrence of readmissions after direct PCI in NSTEMI patients.

Keywords: Machine learning; Non-ST elevation myocardial infarction; PCI; Prediction model; Readmissions.

Figure 1

Study design.NSTEMI: Non-ST Elevation Myocardial Infarction; PCI: Percutaneous Coronary Intervention; LASSO: least absolute shrinkage and selection operator; LR: logistic regression; DT: Decision Tree; RF: Random Forest; SVM: support vector machine; XGBoost: extreme gradient boosting; AdaBoost: Adaptive Boosting; AUC: area under the curve.

Figure 2

Venn plots reflect the number of results for the three variable screening methods; The overlapping part is the seven variables selected, Logistic regression (LR) non-overlapping part has 4 variables, Random Forest (RF) non-overlapping section has 13 variables, Least absolute shrinkage and selection operator (LASSO) has 28 variables in the nonoverlapping section.

Figure 3

ROC for 6 machine learning algorithms. LR: logistic regression; DT: Decision Tree; RF: Random Forest; SVM: support vector machine; XGBoost: extreme gradient boosting; AdaBoost: Adaptive Boosting; AUC: area under the curve.

Figure 4

The nomogram of the LR model; CRP: C reactive protein; TC: Total cholesterol; HDL: high density lipoprotein; LDL: low density lipoprotein.

Figure 5

ROC analysis of optimal LR model and its 7 clinical variables. Outcome: discharge outcomes; mode: admission mode; com: communication ability; CRP: C reactive protein; TC: Total cholesterol; HDL: high density lipoprotein; LDL: low density lipoprotein.

Figure 6

The calibration curve of the LR model. The ideal line indicates that the model prediction is exactly the same as the actual situation, which is the ideal situation. Apparent and bias.

Figure 7

Decision curve analysis of readmission in NSTEMI patients within one year after PCI.Solid black line, assuming no readmissions have occurred in patients (indicated by none, i.e., horizontal line). The grey line indicates that all patients had readmissions (denoted by ALL, i.e., oblique line). The red line represents the calibration curve of the model.

Figure 8

The clinical impact curve of the optimal prediction model is drawn based on the nomogram.

Figure 9

Comparison of ROC curves between the LR prediction model and the adjusted GRACE score, KAMIR score and ACEF score models.