FedAGGEnsemble: a Federated Ensemble Learning based Approach for Malaria Detection

Malaria is one of the life-threatening diseases that is considered as a significant global health challenge, particularly in sub-Saharan Africa. While various machine learning models have been proposed for malaria detection, a critical limitation is their reliance on centralized data, which often compromises patient privacy. To address this challenge, we introduce FedAGGEnsemble, a novel federated ensemble learning framework designed to enhance malaria detection accuracy while safeguarding data confidentiality. By leveraging the power of federated learning, our approach enables collaborative training of multiple local models on decentralized datasets, thereby preserving data privacy. The ensemble technique further improves predictive performance by combining the strengths of diverse models. Our experimental results demonstrate the superiority of FedAGGEnsemble over existing state-of-the-art methods in terms of accuracy and robustness. This research underscores the potential of artificial intelligence to revolutionize public health and highlights the imperative of prioritizing data privacy in medical applications.