Moth-Flame Optimization for Early Prediction of Heart Diseases

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Machine Learning for Disease Prediction

Background:

• Cardiovascular diseases represent a leading global cause of mortality.
• Accurate prediction of heart disease remains a significant challenge in clinical data analysis.
• Existing models for cardiac infection identification often suffer from poor accuracy.

Purpose of the Study:

• To develop a unique method for identifying essential traits for heart disease prediction using machine learning.
• To enhance the effectiveness and accuracy of cardiovascular disease identification.
• To improve the efficiency and reduce the execution time of heart disease classification systems.

Main Methods:

• Utilized ensemble stacking of Decision Tree (DT), Support Vector Machine (SVM), Artificial Neural Network (ANN), and K-Nearest Neighbor (KNN) classification models.
• Integrated the Moth-Flame Optimization (MFO) algorithm for enhanced feature selection.
• Employed logistic regression to create a single best-fit predictive model from the ensemble.

Main Results:

• The proposed system achieved 99% accuracy in predicting heart disease using the Cleveland dataset.
• Feature selection strategies significantly improved classification accuracy and reduced execution time.
• The AI-driven approach demonstrated superior performance compared to existing models.

Conclusions:

• The developed AI-based system offers a highly accurate and efficient method for heart disease prediction.
• The combination of ensemble stacking and MFO optimization enhances diagnostic capabilities.
• This approach holds significant potential for improving clinical decision-making in cardiovascular health.