Title:
CLOUD-INTEGRATED EARLY DETECTION OF ALZHEIMER'S DISEASE USING VGG-CAPSNET

Authors:
Durai Rajesh Natarajan and G. Arulkumaran, India

Abstract:
Alzheimer's disease (AD) progressively erodes the cognitive functions of memory and behavior, establishing it as one standardized dementia of nonelderly persons. Early diagnosis is essential so interventions can occur in time to slow the progression of the illness. Identifying structural damage by MRI scan type is important in diagnosing AD; however, the interpretation is manual-labourintensive, subjective, and error-prone. Progressively, this paper generates a VGG-CapsNet model to detect early stages of Alzheimer's disease using the feature-extraction potential of VGGNet and the spatial hierarchy-preserving ability of Capsule Networks (CapsNets). The model is cloudbased, thus scalable and remotely accessible by healthcare providers. It provides an accurate and efficient diagnosis by using cloud deployment. This increases chances for continuous updates and refined predictions. VGGNet-CapsNet integration above with the cloud infrastructure encapsulates a scalable, efficient, and accurate early detection system for AD, hence helping the healthcare provider classify different stages of the disease.