Enhanced Cardiovascular Disease Prediction Modelling using Machine Learning Techniques: A Focus on CardioVitalnet

Area of Science:

• Cardiology
• Artificial Intelligence
• Data Science

Background:

• Cardiovascular disease (CVD) remains a leading cause of mortality globally.
• Early detection and accurate prediction are crucial for effective intervention and reduced mortality rates.
• Existing predictive models may lack the precision required for personalized risk assessment.

Purpose of the Study:

• To develop an intelligent predictive system for early cardiovascular disease (CVD) detection.
• To integrate deep learning and data mining techniques for enhanced CVD risk prediction.
• To identify optimal features and machine learning models for improved diagnostic accuracy.

Main Methods:

• Data preprocessing and optimized feature selection were performed.
• Deep learning models and machine learning algorithms (Extra Trees, Random Forest, AdaBoost, XG-Boost) were employed for classification.
• Model performance was evaluated using accuracy, sensitivity, and F1-score metrics on a Python platform.

Main Results:

• Machine learning classifiers demonstrated high performance: XG-Boost (99.00%), Random Forest (97.87%), AdaBoost (96.44%), and Extra Trees (94.35%).
• The proposed CardioVitalNet (CVN) achieved an accuracy of 87.45%.
• Feature selection and ML algorithms significantly improved CVD classification effectiveness.

Conclusions:

• Machine learning models, particularly XG-Boost, show significant potential for accurate cardiovascular disease prediction.
• The study provides insights into selecting appropriate models for medical data analysis in CVD risk assessment.
• Optimized feature selection enhances the performance of predictive models for cardiovascular health.