Towards Alzheimer's Disease Classification through Transfer Learning

• DOI: 10.32920/22734329.v1
• Publication Date: 2023-05-03T16:01:01Z
• Authors: Marcia Hon; Naimul Khan
• Keywords: Transfer of learning; Benchmark (surveying); Deep Neural Networks; Training set

<p>Detection of Alzheimer's Disease (AD) from neuroimaging data such as MRI through machine learning have been a subject intense research in recent years. Recent success deep computer vision progressed further. However, common limitations with algorithms are reliance on large number training images, and requirement careful optimization the architecture networks. In this paper, we attempt solving these issues transfer learning, where state-of-the-art architectures VGG Inception initialized pre-trained weights benchmark datasets consisting natural fully-connected layer is re-trained only small images. We employ image entropy to select most informative slices for training. Through experimentation OASIS dataset, show that size almost 10 times smaller than state-of-the-art, reach comparable or even better performance current deep-learning based methods.</p>