Design and Development of Heart Dsisease Prediction System Using Tunicate Swarm Optimized Neural Network

Heart disease is one of the leading causes of death worldwide, emphasizing the urgent need for accurate and early prediction systems. Traditional diagnostic methods often involve manual evaluation and clinical tests, which can be timeconsuming, error-prone, and less accessible in remote settings. While ML models have been used for HD prediction, many still struggle with poor feature selection, overfitting, and limited prediction accuracy. To overcome these limitations, this research introduces a Neural Network-based HD Prediction System optimized using the Tunicate Swarm Optimization (TSO) algorithm. The TSO algorithm, inspired by the collective foraging behavior of tunicates, is employed to optimize the neural network's parameters and enhance feature selection, thereby improving classification performance. The proposed model was evaluated using benchmark HD datasets, achieving a remarkable accuracy of 98.5 %, precision of 97.6 %, recall of 97.8 %, specificity of 98.0 %, and F1-score of 97.7 %. These results demonstrate the system's effectiveness in delivering high diagnostic reliability and reducing false predictions. The integration of TSO with neural networks offers a robust, intelligent, and scalable approach for early HD detection, potentially supporting real-time clinical decision-making.