English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Detection of Alzheimer's Disease (AD) from neuroimaging data such as MRIthrough machine learning has been a subject of intense research in recentyears. Recent success of deep learning in computer vision has progressed suchresearch further. However, common limitations with such algorithms are relianceon a large number of training images, and requirement of careful optimizationof the architecture of deep networks. In this paper, we attempt solving theseissues with transfer learning, where the state-of-the-art VGG architecture isinitialized with pre-trained weights from large benchmark datasets consistingof natural images. The network is then fine-tuned with layer-wise tuning, whereonly a pre-defined group of layers are trained on MRI images. To shrink thetraining data size, we employ image entropy to select the most informativeslices. Through experimentation on the ADNI dataset, we show that with trainingsize of 10 to 20 times smaller than the other contemporary methods, we reachstate-of-the-art performance in AD vs. NC, AD vs. MCI, and MCI vs. NCclassification problems, with a 4% and a 7% increase in accuracy over thestate-of-the-art for AD vs. MCI and MCI vs. NC, respectively. We also providedetailed analysis of the effect of the intelligent training data selectionmethod, changing the training size, and changing the number of layers to befine-tuned. Finally, we provide Class Activation Maps (CAM) that demonstratehow the proposed model focuses on discriminative image regions that areneuropathologically relevant, and can help the healthcare practitioner ininterpreting the model's decision making process.

Transfer Learning with intelligent training data selection for prediction of Alzheimer's Disease