MRI neuroimaging-based Alzheimer’s disease stage classification using deep neural network with convolutional block attention module and GAN-style noise injection

Millions of individuals worldwide suffer from Alzheimer's disease (AD), a chronic, incurable neurological disorder. For the longevity of people, a computer-aided system can contribute to the maximum possible extent. Recently, Deep-learning algorithms have shown better results than machine learning techniques. Researchers have applied CNN models on MRI datasets in various recent studies and have obtained promising results for the early detection of AD. This study proposes a Neuro_CBAM-ADNet diagnostic model for early prediction of the four stages of AD, using MRI digital images. The results show that the proposed model achieved mean accuracy of 98.28%±0.31, which also outperforms the previous related works. The proposed model can identify AD without human intervention and also at an economical cost with high accuracy. The optimistic findings of deep learning models in the early diagnosis of illnesses like AD show that deep learning plays a crucial role in combating these neurological diseases.