Effective prediction of heart disease using hybrid ensemble deep learning and tunicate swarm algorithm

Area of Science:

• Medical Informatics
• Artificial Intelligence in Healthcare
• Cardiovascular Health

Background:

• Heart disease (HD) is a leading global cause of mortality, with increasing prevalence, particularly in India.
• Accurate and timely prediction of heart disease is challenging due to busy lifestyles and subtle symptoms.
• Data mining techniques offer potential for extracting hidden information from clinical datasets for disease prediction.

Purpose of the Study:

• To develop and evaluate a Hybrid Tunicate Swarm Algorithm and Ensemble Deep Learning (TSA-EDL) model for precise heart disease prediction.
• To enhance the accuracy of heart disease diagnosis by integrating advanced data mining techniques.
• To address the critical need for effective heart disease prediction tools in clinical practice.

Main Methods:

• Implementation of a Hybrid TSA-EDL model for heart disease prediction using the Python platform.
• Utilizing DBSCAN (Density-based clustering with noise) for feature selection (identifying relevant, irrelevant, and redundant features).
• Performance evaluation using metrics such as accuracy, recall, specificity, precision, F-score, and error rates on UCI and CVD datasets.

Main Results:

• The proposed Hybrid TSA-EDL model demonstrated superior performance compared to previous algorithms.
• Achieved high accuracy rates of 98.33% on the Cardiovascular Disease (CVD) dataset and 97.5% on the University of California Irvine (UCI) dataset.
• Effective feature grouping and classification were achieved through the integrated approach.

Conclusions:

• The Hybrid TSA-EDL model offers a highly accurate and effective solution for heart disease prediction.
• This approach holds significant potential for early detection and management of heart disease.
• The study highlights the efficacy of combining swarm intelligence algorithms with deep learning for complex medical diagnoses.