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

Self-supervision has been widely explored as a means of addressing the lackof inductive biases in vision transformer architectures, which limitsgeneralisation when networks are trained on small datasets. This is crucial inthe context of cortical imaging, where phenotypes are complex andheterogeneous, but the available datasets are limited in size. This paperbuilds upon recent advancements in translating vision transformers to surfacemeshes and investigates the potential of Masked AutoEncoder (MAE)self-supervision for cortical surface learning. By reconstructing surface datafrom a masked version of the input, the proposed method effectively modelscortical structure to learn strong representations that translate to improvedperformance in downstream tasks. We evaluate our approach on cortical phenotyperegression using the developing Human Connectome Project (dHCP) and demonstratethat pre-training leads to a 26\% improvement in performance, with an 80\%faster convergence, compared to models trained from scratch. Furthermore, weestablish that pre-training vision transformer models on large datasets, suchas the UK Biobank (UKB), enables the acquisition of robust representations forfinetuning in low-data scenarios. Our code and pre-trained models are publiclyavailable at \url{https://github.com/metrics-lab/surface-vision-transformers}.

Surface Masked AutoEncoder: Self-Supervision for Cortical Imaging Data