Prediction and Classification of Alzheimer's Disease Using RESNET-152V2

Alzheimer's disease (AD) is the primary cause of dementia globally and one of the major progressive neurode-generative illnesses, having a major influence on living quality and cognitive function. In order to provide prompt intervention and therapy, as well as possibly halt the disease's course, early diagnosis and precise classification of Alzheimer's disease are essential. This work investigates the use of deep learning methods, more precisely the ResNet-152v2 architecture, for Alzheimer's Disease prediction and classification using neuroimaging data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. To differentiate between people with Alzheimer's disease and those in good health, MRI scans were analysed using the ResNet-152v2 model, which is well-known for its deep and residual learning capabilities. The deep architecture of the model makes it possible to extract intricate features from the imaging data, and residual connections help to improve the model's performance in deep networks by reducing the impact of the vanishing gradient issue. The outcomes of our experiment show that In the categorisation of Alzheimer's disease, the ResNet-152v2 model exhibits high accuracy of 0.94, precision, recall, and F1-scores. These metrics demonstrate how well the algorithm works to minimise false positives and correctly identify patients with Alzheimer's disease. The results indicate that ResNet-152v2 can be a useful tool for the early diagnosis and categorisation of Alzheimer's disease when applied to neuroimaging data. This is a possible method for improving diagnostic accuracy in clinical settings.