Machine Learning-Based Predictive Models for Detection of Cardiovascular Diseases.

Area of Science:

• Cardiology
• Computer Science
• Data Science

Background:

• Cardiovascular diseases pose a significant global health challenge.
• Existing detection methods require advancement, particularly in handling imbalanced datasets which can bias predictions.
• Early detection of myocardial infarction is crucial for improving patient outcomes.

Purpose of the Study:

• To develop accurate and effective early detection methods for heart diseases, focusing on myocardial infarction.
• To address the challenge of imbalanced datasets in cardiovascular disease prediction models.
• To evaluate the performance of various machine learning and deep learning classifiers for heart disease detection.

Main Methods:

• A comprehensive literature review was conducted to identify strategies for handling imbalanced datasets.
• Seven machine learning and deep learning classifiers were deployed: K-Nearest Neighbors, Support Vector Machine, Logistic Regression, Convolutional Neural Network, Gradient Boost, XGBoost, and Random Forest.
• The XGBoost model was meticulously fine-tuned for cardiovascular disease prediction.

Main Results:

• The fine-tuned XGBoost model achieved high performance metrics: 98.50% accuracy, 99.14% precision, 98.29% recall, and 98.71% F1 score.
• The study demonstrated the effectiveness of XGBoost in enhancing diagnostic accuracy for heart disease.
• Performance insights were gained across multiple classifiers for robust prediction model development.

Conclusions:

• Optimized machine learning models, particularly XGBoost, significantly improve the accuracy of early heart disease detection.
• Addressing imbalanced datasets is critical for developing unbiased and reliable cardiovascular disease prediction tools.
• This research provides a strong foundation for developing advanced diagnostic systems for myocardial infarction.