Paper Title
APPLICATIONS OF ARTIFICIAL INTELLIGENCE IN PREDICTIVE HEALTHCARE SYSTEMS

Abstract
India faces a growing cardiovascular disease burden, with over 30 million affected. Diagnosing heart disease is complex, requiring early detection for timely intervention. The vast patient data generated in healthcare makes machine learning (ML) essential for accurate analysis. Traditional diagnosis systems, such as Interval Type-2 Fuzzy Logic System (IT2FLS), suffered from low accuracy and high training time. This research proposes an efficient heart disease prediction system using a Modified Firefly Algorithm (MFA) and Radial Basis Function-Support Vector Machine (RBF-SVM). The dataset comprises input, key, and prediction attributes. Min-max standardization is applied for pre-processing, followed by MFA for handling large datasets. Principal Component Analysis (PCA) reduces redundant features, and RBF-SVM performs classification. To improve accuracy, a Particle Swarm Optimization (PSO) algorithm and RBF-based Trans ductive Support Vector Machine (TSVM) are integrated. Rough set-based attribute reduction using PSO optimizes feature selection, while Opposition-Based Crow Search Optimization (OCSO) further enhances prediction efficiency. Finally, RBF-TSVM performs classification for accurate heart disease prediction. The model’s performance is evaluated using True Positive (TP), True Negative (TN), False Positive (FP), and False Negative (FN) metrics. Results show that the proposed approach outperforms existing methods in accuracy, sensitivity, and specificity. Since heart disease remains a leading global cause of death, predictive models leveraging IoT, cloud computing, machine learning, and deep learning are vital for improving healthcare. Web-based diagnostic systems further aid clinical decision-making, enabling physicians to anticipate and manage cardiac risk factors effectively. Keywords - Heart Disease Prediction, Machine Learning, RBF-SVM, Particle Swarm Optimization (PSO), Feature Selection.