Open Access

REVIEW

Ensemble Deep Learning Approaches in Health Care: A Review

Aziz Alotaibi^*

Department of Computer Science, College of Computing and Information Technology, Taif University, Taif, 21974, Saudi Arabia

* Corresponding Author: Aziz Alotaibi. Email:

(This article belongs to the Special Issue: Advancements in Machine Learning and Artificial Intelligence for Pattern Detection and Predictive Analytics in Healthcare)

Computers, Materials & Continua 2025, 82(3), 3741-3771. https://doi.org/10.32604/cmc.2025.061998

Received 19 September 2024; Accepted 16 January 2025; Issue published 06 March 2025

Abstract

Deep learning algorithms have been rapidly incorporated into many different applications due to the increase in computational power and the availability of massive amounts of data. Recently, both deep learning and ensemble learning have been used to recognize underlying structures and patterns from high-level features to make predictions/decisions. With the growth in popularity of deep learning and ensemble learning algorithms, they have received significant attention from both scientists and the industrial community due to their superior ability to learn features from big data. Ensemble deep learning has exhibited significant performance in enhancing learning generalization through the use of multiple deep learning algorithms. Although ensemble deep learning has large quantities of training parameters, which results in time and space overheads, it performs much better than traditional ensemble learning. Ensemble deep learning has been successfully used in several areas, such as bioinformatics, finance, and health care. In this paper, we review and investigate recent ensemble deep learning algorithms and techniques in health care domains, medical imaging, health care data analytics, genomics, diagnosis, disease prevention, and drug discovery. We cover several widely used deep learning algorithms along with their architectures, including deep neural networks (DNNs), convolutional neural networks (CNNs), recurrent neural networks (RNNs), and generative adversarial networks (GANs). Common healthcare tasks, such as medical imaging, electronic health records, and genomics, are also demonstrated. Furthermore, in this review, the challenges inherent in reducing the burden on the healthcare system are discussed and explored. Finally, future directions and opportunities for enhancing healthcare model performance are discussed.

---

Keywords

---