A likelihood swarm whale optimization based LeNet classifier approach for the prediction and diagnosis of patients...

Area of Science:

• Cardiovascular Medicine
• Artificial Intelligence in Healthcare
• Biomedical Engineering

Background:

• Coronary Artery Disease (CAD), driven by atherosclerosis, represents a global epidemic, contributing significantly to cardiovascular disease (CVD) mortality.
• Atherosclerosis, characterized by arterial hardening and narrowing due to plaque buildup, often progresses asymptomatically, leading to severe outcomes like heart attack and stroke.
• Current diagnostic methods for atherosclerosis lack the sensitivity and resolution for early, non-invasive detection, hindering timely intervention and prevention.

Purpose of the Study:

• To develop and evaluate an intelligent optimization technique for early prediction of atherosclerosis and disease progression.
• To enhance the accuracy and efficiency of diagnosing cardiovascular conditions using advanced computational methods.
• To address the limitations of current diagnostic tools by proposing a novel, non-invasive approach.

Main Methods:

• Utilized the Maximum Likelihood Swarm Whale Optimization (MLSWO) algorithm for feature extraction from Atherosclerosis (STULONG) and Kaggle datasets.
• Employed the LeNet classifier for accurate classification and prediction of disease progression based on extracted features.
• Compared the performance of the proposed MLSWO algorithm against existing state-of-the-art techniques.

Main Results:

• The MLSWO algorithm demonstrated superior performance, achieving an accuracy rate of 99.2%.
• Achieved a high sensitivity rate of 98.36% and a specificity of 100% in disease prediction.
• Outperformed other contemporary methods in identifying crucial features for atherosclerosis prediction.

Conclusions:

• The proposed MLSWO technique offers a highly accurate and effective non-invasive method for early detection and prediction of atherosclerosis.
• This advanced computational approach has the potential to significantly improve the management and prevention of cardiovascular diseases.
• The findings highlight the efficacy of intelligent optimization techniques in enhancing diagnostic capabilities for complex medical conditions.