Deep Convolutional Neural Networks With Ensemble Learning and Generative Adversarial Networks for Alzheimer's Disease Image Data Classification.

Robert Logan,Nicolas Schcolnicov,Akash Indani,Brian G. Williams,Sean J. Miller,Anjali Ganguly,Maria Ferreira da Silva

doi:10.3389/fnagi.2021.720226

Abstract

Recent advancements in deep learning (DL) have made possible new methodologies for analyzing massive datasets with intriguing implications in healthcare. Convolutional neural networks (CNN), which have proven to be successful supervised algorithms for classifying imaging data, are of particular interest in the neuroscience community for their utility in the classification of Alzheimer’s disease (AD). AD is the leading cause of dementia in the aging population. There remains a critical unmet need for early detection of AD pathogenesis based on non-invasive neuroimaging techniques, such as magnetic resonance imaging (MRI) and positron emission tomography (PET). In this comprehensive review, we explore potential interdisciplinary approaches for early detection and provide insight into recent advances on AD classification using 3D CNN architectures for multi-modal PET/MRI data. We also consider the application of generative adversarial networks (GANs) to overcome pitfalls associated with limited data. Finally, we discuss increasing the robustness of CNNs by combining them with ensemble learning (EL).

Highlights

The primary risk factor for developing Alzheimer’s disease (AD) is advanced age (Hebert et al, 2010)
In most Convolutional neural networks (CNN) architectures, the output of the convolutional layer is passed through a non-linear activation function as in a traditional neural network
Prakash et al (2019) tested the GoogLeNet, AlexNet, and VGG-16 networks for classification of the Alzheimer’s Disease Neuroimaging Initiative (ADNI) magnetic resonance imaging (MRI) dataset in cognitively normal (CN), mild cognitive impairment (MCI), and AD to show that GoogLeNet achieves 99.84% accuracy in training and 98.25% in the test set, higher than the remaining architectures

Summary

INTRODUCTION

The primary risk factor for developing Alzheimer’s disease (AD) is advanced age (Hebert et al, 2010). A commonly used resource in studying AD imaging data with the application of deep learning is the Alzheimer’s Disease Neuroimaging Initiative (ADNI) reference dataset. This neuroimaging database includes data from AD, mild cognitive impairment (MCI), and healthy individuals (Petersen et al, 2010). It consists of over 50,000 patient images and is often used to test the performance of models in AD image classification. We discuss recent exciting advances in the deep learning analysis of neuroimaging for AD patient diagnostics

BACKGROUND

Findings

Method