A machine learning model for predicting short-term in-hospital mortality in acute myocardial infarction with coexisting chronic obstructive pulmonary disease
By
Weibin He
Jieli Sheng
Shuxiong Cai
Lihong Zheng
Zhenzhao Wang
Shujiao Zheng
Xinqi Lai
Chun Yang
Yiting Ke
Xiaohong Huang
June 25, 2026
Clinical Scorecard: A predictive machine learning framework for assessing short-term in-hospital mortality in acute myocardial infarction patients with concurrent chronic obstructive pulmonary disease
At a Glance
Category Detail
Condition Acute Myocardial Infarction with Chronic Obstructive Pulmonary Disease Key Mechanisms Systemic inflammation and oxidative stress linking COPD and AMI. Target Population ICU patients with acute myocardial infarction and chronic obstructive pulmonary disease. Care Setting Intensive Care Unit
Key Highlights
Logistic regression model showed superior discriminative performance with an AUC of 0.782. 28-day in-hospital mortality rate was 27.9% among the studied population. Eight key variables identified for mortality prediction include age, heart rate, and lactate dehydrogenase. Guideline-Based Recommendations
Diagnosis
Use machine learning models to assess risk in patients with AMI and COPD. Management
Implement early identification of high-risk patients to guide treatment decisions. Monitoring & Follow-up
Regularly assess key predictors such as heart rate and lactate levels. Risks
Patients with COPD and AMI face significantly elevated mortality risks. Patient & Prescribing Data
ICU patients with acute myocardial infarction and chronic obstructive pulmonary disease.
Consider the use of β-blockers and ACE inhibitors/ARBs in management.
Clinical Best Practices
Utilize predictive models to enhance clinical decision-making. Incorporate SHAP analysis for better interpretability of model predictions. Related Resources & Content
