Federated learning with integration of decision making method and various machine learning algorithms for Alzheimer's prediction

Alzheimer's disease, a progressive and debilitating neurodegenerative disorder, presents considerable challenges in early diagnosis and treatment planning. Given the sensitive nature of patient health records and the diversity of medical data sources, there is a pressing need for diagnostic tools that are not only accurate and robust but also privacy-preserving. Federated learning offers a promising solution by enabling collaborative model training across multiple decentralized institutions, allowing each to contribute to a shared global model without exposing raw data. This approach safeguards patient confidentiality while ensuring compliance with data protection regulations. To further enhance the efficiency and effectiveness of federated systems, this research integrates multi-criteria decision-making methods into the federated learning framework. The use of these facilitates informed client selection, balanced model aggregation, and the prioritization of key factors such as accuracy, data distribution, volume, and computational capacity. This integration enables performance-driven decision-making under heterogeneous data conditions and enhances the scalability and personalization of collaborative learning. Various machine learning algorithms are incorporated within this federated decision-making framework to evaluate client contributions, optimize model training, and ensure the selection of top-performing clients based on multiple criteria. These algorithms play a crucial role in constructing accurate, robust and privacy-preserving models across distributed data sources, enabling effective collaboration without compromising data privacy. Together, federated learning and decision-making methods form a powerful paradigm for building intelligent, secure and high-performance diagnostic systems tailored to the complexities of Alzheimer's disease. This research study provides an in-depth exploration of the working mechanism of federated learning combined with decision-making method. It includes the evaluation metrics calculated by these methods, a comparison of various machine learning algorithms utilized in this federated learning decision-making framework, as well as a discussion of their limitations and future directions.