Tools
Tools
Ahamed, B. Shamreen, Virdee, Bal Singh, Khanna, Ashish, Vidhya, R. G. and Sivashankar, S. (2026) Hybrid deep ensemble learning with metaheuristic optimization for heart disease prediction. In: 2026 Fourth International Conference on Augmented Intelligence and Sustainable Systems (ICAISS), 28 - 30 April 2026, Trichy, India.
The major causes of deaths due to heart disease rank among the top for global health concerns, with predicting heart disease for diabetic patients a major challenge. This research study proposes a deep ensemble learning with metaheuristic optimization framework for heart disease prediction. The workflow of this model includes performing extensive preprocessing of the provided dataset, followed by an optimization algorithm-based feature selection. Furthermore, the proposed model utilizes a weighted fusion method for combining various health data sources, convolution-pooling, attention, and spatial-temporal learning. Ensemble stacking learning is also used with an SVM learner, with Pufferfish Optimization Algorithm used for fine-tuning. As indicated, the proposed model was successful in achieving 0.980 accuracy, 0.970 precision, 0.990 recall, and 0.950 F1-score for Dataset 1, with 0.990 accuracy, 0.980 precision, 0.970 recall, and 0.980 F1-score for the Cleveland dataset.
Available under License Creative Commons Attribution 4.0.
Download (613kB) | Preview
 |
View Item |