Agglomeration of deep learning networks for classifying binary and multiclass classifications using 3D MRI images for early diagnosis of Alzheimer’s disease: a feature-node approach

Abstract

Alzheimer’s disease is a degenerative brain con dition causing memory loss in the elderly. Existing machine learning methods often yield low classifcation accuracy due to evaluating single modality features. This paper presents a novel approach that combines Graph Attention Networks and Deep Convolutional Graph Neural Networks to leverage 3D 1.5 T and 3 T T1-weighted MRI images as nodes, ena bling faster feature extraction. Three Graph Convolutional Network layers are introduced to improve the classifcation accuracy for three binary classifcations (AD vs. CN, MCI vs. CN, and MCI vs. AD) and multiclass classifcation (AD vs. CN vs. MCI). The model is optimized for weight updates using the Adaptive Stochastic Gradient Descent technique. Comparative analysis with efcient 3DNET, Squeeze3D NET, and GoogLENET demonstrates superior performance of the proposed DCGNN network. Furthermore, evaluations against four state-of-the-art techniques for binary and mul ticlass classifcations show its potential in diagnosing the early stages of Alzheimer’s disease. The developed model exhibits promise as an efective tool for diagnosing Alzhei mer’s disease at its early stages.