Walrus Optimization Algorithm based Support Vector Machine for Feature Selection and Classification in Coronary Artery Disease

Coronary artery disease (CAD) is still one of the top causes of death globally, highlighting the need for precise diagnostic methods. Machine learning (ML) has become a valuable tool in medical diagnostics, leveraging vast data to enhance disease detection. However, the success of ML models in CAD classification depends heavily on feature selection to mitigate the effects of redundant and irrelevant data attributes. In proposed manuscript, a new method is introduced to choose the most useful features for detecting CAD using Walrus Optimization Algorithm (WaOA). Proposed method effectively search through all the possible features and identify the ones that are most helpful for accurately diagnosing CAD. The WaOA is integrated with Support Vector Machine (SVM) classifier, which uses the selected features to classify dataset as either having CAD or not. By combining these two techniques, we can improve the overall accuracy of the CAD classification model. The proposed method is compared with several algorithms and found that it performed better than existing approaches on real-world CAD data.