Enhanced cardiovascular disease classification using the mayfly algorithm and real-time data

Abstract

Cardiovascular diseases are the leading cause of death worldwide, therefore precise and timely diagnosis improves patient outcomes while lowering medical costs. Effectiveness of classification models is still limited by the challenges associated with managing high-dimensional medical datasets, despite advancements in machine learning. Traditional feature selection strategies are often ineffective due to the data dimensionality and complexity. This study improves feature selection and classification accuracy in cardiovascular disease datasets by utilising the Mayfly Algorithm (MA), a novel meta-heuristic optimisation technique inspired by mayfly mating behaviour. The MA is used in the study to find the best features using five real-time cardiovascular datasets. The chosen features are evaluated using classifiers such as Random Forest (RF), K-Nearest Neighbours (KNN), and Support Vector Machine (SVM). Traditional optimization techniques, including Genetic Algorithm (GA), Particle Swarm Optimisation (PSO), and Ant Colony Optimisation (ACO), are examined and contrasted. With a large reduction in feature space and good classification accuracy, the MA showed superior performance in feature selection. Compared to previous approaches, which varied between 80% and 85%, the classification accuracies projected to be attained are in the range of 90% to 95% across the five datasets. Significant gains in classification accuracy and feature reduction were attained by the MA’s consistent selection of the most pertinent features. The Mayfly Algorithm’s potential for medical data analysis is demonstrated by this work, especially for high-dimensional cardiovascular datasets. When it comes to real-time disease classification, MA outperforms other optimization strategies in terms of accuracy and effectiveness. The higher accuracy and reduced feature set demonstrate the efficacy of machine learning, pointing to potential wider uses in medical diagnostics.