You should be logged in to see the presentation
Diffusion Tensor Imaging (DTI) and Magnetic Resonance Imaging (MRI) techniques have gained significant popularity in the diagnosis of neurodegenerative disorders. Combining brain scans with deep learning is receiving increasing attention in medical diagnostic applications. However, deep networks can learn powerful features and perform well only when a large amount of DTI or MRI image data are available. The paper is based on the concept that in Alzheimer's Disease (AD) white matter tracts of the brain are affected earlier than gray matter and attempts to reduce the dependence on massive training data by exploiting transfer learning of deep networks pretrained on ImageNet data. The proposed approach includes MRI and DTI data preprocessing and performs classification of Mild Cognitive Impairment (MCI), AD and normal patient data with transfer learning that uses modifications of the AlexNet and VGG16 convolutional neural networks (CNNs). Experiments using data from the ADNI database demonstrate the advantage of DTI trained models in the prediction of Alzheimer’s disease (AD). At the same time, MRI trained models appear to perform better when detecting MCI. The highest accuracy of 99.75% in the diagnosis of AD was achieved with VGG models using DTI scans. The prediction of early cognitive decline with an accuracy of 93% was reached by VGG models processing MRI data.
Birkbeck, University of London, United Kingdom
There are still no questions. Be the first to post one!
You should be logged in order to use this feature.