Distinguishing Alzheimer's disease using SPM toolbox and deep learning techniques

In this paper, we use Convolutional neural networks to detect Alzheimer's disease (AD) from mild cognitive impairment (MCI) and Normal Control (NC). Alzheimer's disease (AD), The most common cause of dementia in the elderly,is aprogressive neurodegenerative disorder that gradually robs the patient of cognitive function and eventually causes death. In recent years, with the increase in life expectancy globally, diagnosing AD has become very important. If MCI develops, the patient's mental abilities are irreversibly impaired, leading to Alzheimer's disease and dementia. This disorder has received special attention from many researchers; Because by diagnosing it in the early stages, its progression can be stopped, and treatment can be taken. Common ways to diagnose the disease are biochemical tests and psychological tests. One of the proposed approaches for diagnosing Alzheimer's disease is the analysis of Magnetic resonance imaging (MRI) used to study changes in the structure of the human brain. In this paper, brain magnetic resonance images (MRI) are pre-processed using the SPM toolbox, then the brain's gray matter (GM) is segmented and given as input to the CNN algorithm. This article uses the ADNI dataset. The results of this test show that we were able to classify the three categories of standard control (NC), Alzheimer’s disease (AD), and mild cognitive impairment (MCI) With an accuracy of over ۹۹%.