Key result
Novel deep learning model outperforms standard machine learning with an AUC of 0.99 for MI prediction.
Why the study?
Myocardial infarction is a leading cause of morbidity and mortality worldwide, complicated by intricate interactions among physiological parameters, comorbidities, and lifestyle habits.
Does a deep learning-based predictive model incorporating physiological, comorbidity, and lifestyle data improve myocardial infarction risk forecasting in a large, imbalanced patient cohort?
Does a deep learning-based predictive model incorporating physiological, comorbidity, and lifestyle data improve myocardial infarction risk forecasting in a large, imbalanced patient cohort?
A deep learning framework incorporating clinical, physiological, and lifestyle factors with SMOTE and SHAP can effectively identify high-risk myocardial infarction patients in highly imbalanced real-world datasets.
Answers are grounded in this paper and cite their sources.
Deep learning models aid MI risk stratification in large cohorts; leaves open prospective validation before clinical use.
Myocardial infarction is one of the most common causes of morbidity and mortality across the globe, complicated by intricate associations between physiological parameters, comorbidities and lifestyle habits. In the proposed study, a large-scale cardiovascular dataset of 281,934 patient records having 35 clinical features was used by the deep learning–based models for risk prediction. Random forest (RF), Naive Bayes, TabNet and one-class support vector machine were evaluated. Incorporating nonlinear feature transformation by the exponential linear unit (ELU) prior to RF classification (ELU on RF), with better accuracy, categorizes the normal and high-risk patients. Investigations were carried out to evaluate whether the root cause was an imbalanced dataset. The synthetic minority oversampling technique worked on the severe imbalance of the dataset, and Shapley additive explanations and LIME were helpful in finding a target column. Results suggested early prediction of the risk and bringing evidence-based preventive care.
No takes yet. Share an insight, caveat, or question.
Sarkar et al. (2026) studied Myocardial Infarction Risk (n=281,934). Exponential Linear Unit (ELU) on Random Forest (RF) model vs. Standard Random Forest, TabNet, and Naive Bayes models was evaluated on Area Under the Curve (AUC) for myocardial infarction risk prediction. The proposed Exponential Linear Unit on Random Forest model achieved an Area Under the Curve of 0.99 for predicting myocardial infarction risk, outperforming standard machine learning models.
Synapse has enriched 10 related papers. Consider them for comparative context: