Convolutional Neural Networks Algorithm for Detecting Alzheimer's Disease

The identification of Alzheimer's disease (AD) has become crucial in recent years due to the global increase in life expectancy. If mild cognitive impairment (MCI) occurs, it can progress to Alzheimer's disease and dementia because it permanently impairs the patient's mental ability. Many researchers have given this condition their undivided focus since, if caught early enough, it can be treated and its progression halted. Psychological examinations and biochemical tests are frequently used to diagnose the illness. The analysis of magnetic resonance imaging (MRI) scans, which are used to examine changes in the structure of the human brain, is one of the suggested methods for detecting Alzheimer's disease. The SPM (Statistical Parametric Mapping) toolbox is used in this study to preprocess brain MRI images before segmenting the brain's gray matter (GM) and feeding it into the convolutional neural network (CNN) algorithm. The ADNI (Alzheimer's Disease Neuroimaging Initiative) dataset is used in this paper. Based on the test's results, we could accurately distinguish the three groups of normal control (NC), Alzheimer's disease, and moderate cognitive impairment.