Investigation of Various Transfer Learning Techniques for Classifying Alzheimer's Disease Dataset

Preview this chapter:

Investigation of Various Transfer Learning Techniques for Classifying Alzheimer's Disease Dataset, Page 1 of 1

< Previous page | Next page > /docserver/preview/fulltext/9789815274134/chapter-16-1.gif

Alzheimer's disease, also called dementia, is a severe psychological disorder that affects the cerebrum, causes cognitive decline, and impairs a person's ability to reason. During the initial stage, AD patients encounter the standard adverse effects. Usually, they lose track of their everyday tasks and obligations. They find it challenging to communicate verbally with people. They also have a diminished ability to think critically, among other things. Currently, research is ongoing to determine the underlying cause of AD. We are now quite concerned about AD because the majority of its patients are older than sixty. Nervous system specialists typically perform multiple tests to differentiate AD. Human errors occur from time to time. We require high-performing deep-learning models to diagnose and forecast the illness better. This study looks at how well VGG16, InceptionResNet-V2, Resnet50, Resnet101, and Resnet152 classified the AD dataset and compares and contrasts their results. Accuracy, loss, validation accuracy, and validation loss were the performance indicators we utilized to assess the models. You may find the dataset on the Kaggle repository.