Alzheimer’s Disease Multi-class Classification Model Based on CNN and StackNet Using Brain MRI Data

Abstract

AbstractAlzheimer's Disease (AD) may harm memory cells forever, which results in dementia. The detection and classification of Alzheimer's disease (AD) are critical in patient care. Many studies have applied machine learning and deep learning methods to classify the stages of AD. Consequently, it is used for the grouping of cerebrum images among Non-Demented, Very Mild Dementia, Mild dementia, Moderate Dementia, Alzheimer's Disease (AD) which are the four classes of AD in this manner guaranteeing exact and precise diagnosis. In clinical examination, magnetic resonance imaging (MRI) is utilized to analyze AD. For exact classification of dementia stages, we need profoundly discriminative high-lights acquired from MRI images. In this paper, we proposed a new model based on three convolution neural network architectures, DenseNet196, VGG16 and ResNet50 pre-trained models for features extraction, and the stacking ensemble for multi-class classification of this disease using a brain MRI dataset. This model has achieved an accuracy of 89% using the dataset published on Kaggle.KeywordsAlzheimer’s diseaseConvolution neural networkMachine learn ingDeep learningStackingMRI dataset