Cardiovascular Disease Prediction Using Super Learner

Abstract

This project addresses the global health challenge presented by cardiovascular disease (CVD), with a specific focus on Ischaemic Heart Disease (IHD), commonly known as coronary heart disease (CHD). CHD involves the narrowing of coronary arteries due to arterial plaque buildup, contributing significantly to substantial mortality rates worldwide. The project recognizes the importance of early and accurate detection of CVD, as demonstrated by clinical studies, to improve patient survival rates.However, barriers such as the high cost of diagnosis and the financial burden of treating the disease hinder effective healthcare delivery. Existing studies often oversimplify CHD classifications, overlooking the full range of severity levels within the disease. This study seeks to overcome these limitations by employing Machine Learning (ML) algorithms, including Random Forest Classifier (RFC), Support Vector Machine (SVM), Multi-Layer Perceptron (MLP), etc, within an ML ensemble known as the Super Learner.The research focuses on the urgency to accurately categorize patients into specific severity levels, optimizing investigation time and cost. The ML ensemble, Super Learner, combines diverse base learners to create a model that surpasses individual models, providing robust predictions across diverse scenarios. The achievements of the project include the development of a predictive model with the ability to classify CHD beyond binary classifications, achieving an unprecedented ROC score of 0.96. This performance underscores the model's potential as a valuable tool in the early diagnosis and management of CHD.

Note:

Funding Information: Sheffield Hallam University.

Conflict of Interests: We, the authors, declare that we have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper

Keywords: Machine Learning, Super Learner, UCI Dataset Repository, Cardiovascular Disease