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

Neuroimaging of large populations is valuable to identify factors thatpromote or resist brain disease, and to assist diagnosis, subtyping, andprognosis. Data-driven models such as convolutional neural networks (CNNs) haveincreasingly been applied to brain images to perform diagnostic and prognostictasks by learning robust features. Vision transformers (ViT) - a new class ofdeep learning architectures - have emerged in recent years as an alternative toCNNs for several computer vision applications. Here we tested variants of theViT architecture for a range of desired neuroimaging downstream tasks based ondifficulty, in this case for sex and Alzheimer's disease (AD) classificationbased on 3D brain MRI. In our experiments, two vision transformer architecturevariants achieved an AUC of 0.987 for sex and 0.892 for AD classification,respectively. We independently evaluated our models on data from two benchmarkAD datasets. We achieved a performance boost of 5% and 9-10% upon fine-tuningvision transformer models pre-trained on synthetic (generated by a latentdiffusion model) and real MRI scans, respectively. Our main contributionsinclude testing the effects of different ViT training strategies includingpre-training, data augmentation and learning rate warm-ups followed byannealing, as pertaining to the neuroimaging domain. These techniques areessential for training ViT-like models for neuroimaging applications wheretraining data is usually limited. We also analyzed the effect of the amount oftraining data utilized on the test-time performance of the ViT via data-modelscaling curves.

Efficiently Training Vision Transformers on Structural MRI Scans for Alzheimer's Disease Detection