Efficient Diagnosis of Coronary Artery Disease: A Hybrid Convolutional-XGBoost Approach

Cardio Vascular Diseases (CVDs) are the prime reason of death globally. Particularly, the cases of coronary artery disease (CAD) have been rising in the post Covid era, especially amongst middle-aged people. Thus, the before time detection of the disease might evade the growing number of fatalities and morbidities. Two most common and accurate methods used to detect CAD are angiography and electrocardiography (ECG). While the former is invasive and a risky procedure, the latter one although non-invasive lacks correctness. Over a decade, non-invasive Machine Learning (ML) as well as Deep Learning (DL) methods have shaped the healthcare scenario by facilitating early diagnosis of the diseases through the application of various computer based algorithmic models over clinical data and images. A hybrid Convolutional- Extreme Gradient Boosting (Convolutional XGBoost) model has been proposed to accurately detect CAD by outperforming some of the existing methods.