Fair and explainable Myocardial Infarction (MI) prediction: Novel strategies for feature selection and class imbalance correction

Keyphrases

• Myocardial Infarction 100%
• Feature Selection 100%
• Early Detection 100%
• Imbalanced Data 100%
• Class Imbalance Mitigation 100%
• Feature Class 100%
• Dual-path 100%
• Myocardial Infarction Prediction 100%
• Myocardium at Risk 100%
• National Health Interview Survey 50%
• Healthcare Professionals 50%
• Risk Factors 50%
• Health Surveys 50%
• Feature Relevance 50%
• Behavioral Risk Factor Surveillance System (BRFSS) 50%
• Artificial Neural Network 50%
• Electronic Health Records 50%
• Model-driven Development 50%
• Innovative Technique 50%
• Predictive Factors 50%
• Risk Prediction 50%
• Artificial Neural Network Model 50%
• Shapley Additive Explanations 50%
• Coronary Heart Disease 50%
• Balanced Performance 50%
• Prediction Difference 50%
• Criteria Selection 50%
• ROC-AUC 50%
• Class Imbalance 50%
• Prediction Bias 50%
• Improved Patient Outcomes 50%
• Bronchiolitis 50%
• Weighted Sampling 50%
• Rising Incidence 50%
• Survey Datasets 50%
• Non-modifiable Risk Factors 50%
• Minority Class 50%
• Reliable Model 50%
• Personal Health Information 50%
• Biased Decision-making 50%

Computer Science

• Feature Selection 100%
• Early Detection 100%
• Feature Extraction 100%
• Artificial Neural Network 100%
• Class Imbalance 100%
• Neural Network Model 50%
• Generalizability 50%
• Model Development 50%
• SHapley Additive exPlanation 50%
• Decision Making 50%
• Selection Criterion 50%

Medicine and Dentistry

• Myocardial Infarction 100%
• Feature Extraction 100%
• Decision Making 16%
• Health Survey 16%
• Apoplexy 16%
• Sensitivity and Specificity 16%
• Behavioral Risk Factor Surveillance System 16%
• Ischemic Heart Disease 16%
• Electronic Health Record 16%
• Traditional Risk Factor 16%
• Bronchitis 16%