Clinical Data Analysis for Prediction of Cardiovascular Disease Using Machine Learning Techniques

Area of Science:

• Cardiology
• Computer Science
• Data Science

Background:

• Cardiovascular disease (CVD) detection is challenging due to risk factors like hypertension and high cholesterol.
• Accurate risk assessment and treatment are crucial for managing cardiac conditions.
• Advancements in machine learning (ML) are poised to transform clinical healthcare practices.

Purpose of the Study:

• To develop and recommend a highly accurate machine learning-based system for cardiovascular disease prediction.
• To identify the optimal ML model for classifying cardiovascular datasets.

Main Methods:

• Utilized several modern machine learning algorithms: REP Tree, M5P Tree, Random Tree, Linear Regression, Naive Bayes, J48, and JRIP.
• Applied these algorithms to classify popular cardiovascular datasets.
• Evaluated model performance using various metrics to determine the best predictive model.

Main Results:

• The Random Tree model demonstrated superior performance in predicting cardiovascular disease.
• Achieved 100% accuracy, lowest Mean Absolute Error (MAE) of 0.0011, and lowest Root Mean Square Error (RMSE) of 0.0231.
• The Random Tree model offered the fastest prediction time at 0.01 seconds.

Conclusions:

• The Random Tree algorithm is highly effective for cardiovascular disease prediction.
• The proposed Cardiovascular Disease Prediction System (CDPS) offers a reliable and efficient tool for clinical use.
• Machine learning holds significant potential for improving early detection and management of cardiac conditions.