Identifying and Categorizing Alzheimer's Disease with Lightweight Federated Learning Using Identically Distributed Images

Among the leading causes of mortality in developed countries are neurological diseases like Alzheimer's disease (AD). Amazing results have come out of studies that have used algorithms powered by computers. Deep models are all the rage right now, especially in the world of image processing. Many academic studies have focused on deep learning for Alzheimer's disease detection since 2018. This paper introduces a simplified framework for federated deep learning to classify Alzheimer's diseases while ensuring the security of private information. This work uses IID (Independent and Identically Distributed) to test federated deep learning models using a distributed client-server architecture to safeguard client data. The framework was tested with one and numerous clients in traditional and federated learning situations. AlexNet's accuracy of 98.53 percent made it our standard after analyzing other pre-trained models. IID datasets were then evaluated using federal learning (FL). The framework's advanced features make it the optimal choice for early-stage diagnosis and classification of Alzheimer's disease.