Comprehensive Evaluation of Federated Learning Based Models for Disease Detection in Healthcare

In healthcare, the primary objective is to improve the individuals' well-being by addressing a range of health issues through accurate disease detection. Central to this process is Federated Learning (FL), a novel Machine Learning (ML) technique which aids model training across distributed devices, ensuring data privacy. FL, in conjunction with Artificial Learning (AI) algorithms, enhances disease detection, treatment, and management practices, ultimately improving patient outcomes and healthcare services. This study analyzes 50 research papers on FL and AI-based healthcare disease detection to assess existing models comprehensively. Through meticulous literature review and gap analysis, the review aims to provide insights and recommendations for robust FL-based solutions. Exploring FL's potential, challenges, and implications in healthcare, the review aims to stimulate the development of AI-assisted disease detection systems. This work involves systematic paper selection, categorization, and in-depth review, followed by chronological trend analysis, dataset review, and performance metric evaluation. The study also identifies common limitations in healthcare disease detection using AI algorithms, underlining the need for efficient FL-based approaches.