Authors :

Dr. K. Sampathkumar AND S. Periyasamy

Address :

Guest Lecturer, Department of Mathematics, Government Arts College (Grade-1), Ariyalur, India

Guest Lecturer, Department of Computer Science, Government Arts College, Harur, India, [email protected]

Abstract :

The significant component of the body part is heart function leads to human life prediction. Heart failure may be inactive, but it can be predicted through health screenings and the individual's daily activities. The contemporary healthcare system is being revolutionized by Deep Learning (DL) concepts, yet challenge for forecasting accurate cardiac illness and reliability. With medical data, it is hoped that combining the two procedures will improve the effectiveness of established strategies for predicting heart disease. The research that is being suggested presents a novel way for boosting heart illness forecasting preciseness while simultaneously integrating with nature-inspired optimization techniques that are associated with existing feature engineering methods. Hence, the combined approach of Binary Particle Swarm Optimization, assisted with Attention-based Deep Network (BPSO-ADN), has been suggested to efficiently extract significant features from the dataset to enhance prediction accuracy. The research uses a cardiac dataset followed by pre-processing, which needs to be considered. BPSO, used for feature selection, and ADN, used for analyzing detailed patterns within the data, provide a synergistic relationship at the heart of the technique. Throughout its repeated exploration of feature subsets, BPSO is directed by a fitness evaluation method to arrive at the subset most suitable for the prediction of heart disease. By utilizing self-attention processes, the ADN component can identify dependencies within the data, which ultimately results in higher predictive performance for both binary and multivariate classification of attributes.

Keywords :

Binary Particle Swarm Optimization; Deep Neural Network; Heart Disease Prediction: NatureInspired Optimization; Self-Attention Model.

DOI :