Classification of Alzheimer’s Disease Using 2D/3D Convolutional Neural Networks
Kiran, Bidare Divakarachari Parameshachari, D S Sunil Kumar, K V Sudheesh, R Deepak, H A Deepak · 2023
The neurodegenerative brain disorder known as Alzheimer’s disease (AD) is brought on by the buildup of amyloid proteins, the formation of plaques, and the loss of neurons. Parkinson’s disease (PD), another prevalent subtype of dementia like AD, is characterized by the loss of dopaminergic neurons in the midbrain region known as the substantia nigra pars compacta. Both AD and PD aim to reduce healthcare expenses by reducing the world’s ageing population. Therefore, techniques that aid in the early detection of these disorders are required. In the area of medical imaging, where there is typically a limited amount of data that can be used for training, these drawbacks make it difficult to employ the most recent deep learning methods. Humans have difficulty distinguishing between early forms of Alzheimer’s disease and correctly diagnosing them. Convolutional networks can be used to classify human MRI scans, as this paper demonstrates. Mild Cognitive Impairment, Alzheimer’s Disease, and Clinically Normal are the three categories. Finally, a functional model constructed with three-dimensional convolutional layers that performs well across the three classes is presented. The assessment results show that the proposed strategy accomplishes the model exhibition on Promotion conclusion by normal of 82%exactness.