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

Introspection of deep supervised predictive models trained on functional andstructural brain imaging may uncover novel markers of Alzheimer's disease (AD).However, supervised training is prone to learning from spurious features(shortcut learning) impairing its value in the discovery process. Deepunsupervised and, recently, contrastive self-supervised approaches, not biasedto classification, are better candidates for the task. Their multimodal optionsspecifically offer additional regularization via modality interactions. In thispaper, we introduce a way to exhaustively consider multimodal architectures forcontrastive self-supervised fusion of fMRI and MRI of AD patients and controls.We show that this multimodal fusion results in representations that improve theresults of the downstream classification for both modalities. We investigatethe fused self-supervised features projected into the brain space and introducea numerically stable way to do so.

On self-supervised multi-modal representation learning: An application to Alzheimer's disease