AlzheimerNet: Deep Learning for Alzheimer’s Stage Classification from MRI

Abstract

Alzheimer's disease (AD) is a prevalent neurodegenerative condition that has serious consequences for patient care and public health. Early detection and appropriate staging of Alzheimer's disease are critical for effective treatment planning and disease management. In this paper, we introduce AlzheimerNet, a novel deep learning framework designed for AD stage classification using Magnetic resonance imaging (MRI) Using functional brain changes recorded in MRI scans; AlzheimerNet demonstrates promising results in accurately categorizing patients into different stages of AD progression. The input to AlzheimerNet is a pre-processed MRI scan, which undergoes a series of convolutions to extract spatial features representing structural abnormalities associated with AD. These features are then fed into fully connected layers, which further encode the extracted information to facilitate stage classification. The final layer of the network utilizes a softmax activation function to output probabilities corresponding to different AD stages, allowing for multi-class classification. This paper presents the architecture, implementation details, and experimental results of AlzheimerNet, highlighting its potential impact on better diagnosing and managing Alzheimer's disease