Revolutionizing Disease Diagnosis: A Microservices-Based Architecture for Privacy-Preserving and Efficient IoT Data Analytics Using Federated Learning

• DOI: 10.48550/arxiv.2308.14017
• Publication Date: 2023-01-01
• Authors: Safa Ben Atitallah; Maha Driss; Henda Ben Ghézala
• Keywords: Microservices; Data Analysis; Differential Privacy

Deep learning-based disease diagnosis applications are essential for accurate at various stages. However, using personal data exposes traditional centralized learning systems to privacy concerns. On the other hand, by positioning processing resources closer device and enabling more effective analyses, a distributed computing paradigm has potential revolutionize diagnosis. Scalable architectures analytics also crucial in healthcare, where results must have low latency high dependability reliability. This study proposes microservices-based approach IoT satisfy performance requirements arranging entities into fine-grained, loosely connected, reusable collections. Our relies on federated learning, which can increase accuracy while protecting privacy. Additionally, we employ transfer obtain efficient models. Using than 5800 chest X-ray images pneumonia detection from publicly available dataset, ran experiments assess effectiveness of our approach. reveal that performs better identifying cutting-edge technologies, demonstrating approach's promising performance.