Deep Learning-Based Ensembling Technique to Classify Alzheimer’s Disease Stages Using Functional MRI

Table 6

Comparison of state of the methods for AD classification.

Study	Year	Stages	Method	Ensembling technique	Accuracy
Loddo et al. [22]	2022	NC/VM-AD/Mi-AD/Mo-AD	AlexNet, ResNet-101, and Inception ResNetV2	Averaging	98.24%
Fang et al. [23]	2020	AD/HC, MCI/HC	GoogleNet, ResNet, and DenseNet	AdaBoost	93%
Wang et al. [11]	2020	AD/MCI/CN	DenseNet	Probability-based fusion	97.52%
Karwath et al. [24]	2017	AD/healthy, mild MCI/severe MCI	Alexnet CNN	Majority voting	AD/Healthy => 91%,; Mild MCI/severe MCI => 85%
This study	2022	AD/SMC/EMCI/MCI/LMCI/CN	VGG-16, ResNet-18, Inception V1, AlexNet, and Custom CNN	Max voting	98.8%