analiza aiqbench
This research introduces vSV-ViT, a Vision Transformer architecture specifically engineered to handle the non-Euclidean geometry of 3D brain cortical surfaces by utilizing variable-size supervertex partitioning. By addressing limitations in existing fixed-patch methods that often conflate anatomical regions or include non-cortical vertices, the model demonstrates superior predictive accuracy for Alzheimer's disease classification and sex prediction tasks. The work advances machine learning techniques for processing complex medical imaging data, offering a robust framework for future applications in neuroimaging analysis.
vSV-ViT: Variable-size SuperVertex Vision Transformer for Cortical Surface Learning in Alzheimer's Disease
arXiv:2605.26514v2 Announce Type: replace-cross Abstract: Learning from 3D meshes is challenging because the data reside on non-Euclidean surfaces embedded in 3D space. This is particularly evident in domains such as brain cortical surface analysis, where existing models typically rely on ROI-agnostic, face-based, or fixed-size patches. Such patches can duplicate boundary vertices, conflate anatomically distinct regions, or incorporate non-cortical vertices such as those of the medial wall. We develop variable-size supervertex (vSV) partitioning, which groups cortical surface vertices into vSV patches similar to superpixels in images. Building on these vSVs, we design Variable-size SuperVertex Vision Transformer (vSV-ViT), a Vision Transformer that processes vSVs through padding, mask-aware patch embedding, and vertex position embedding. We demonstrate the benefits of the proposed framework through a quantitative analysis of vertex assignment in surface partitioning and downstream tasks using Alzheimer's disease brain image datasets. For downstream tasks, we use meshes of cortical thickness and cortical curvature as input to various encoder models and evaluate classification performance. The superior predictive accuracy of vSV-ViT over recent surface-based models supports its effectiveness for cortical surface learning, with AD-related and sex prediction serving as downstream validation. Extending the framework to other cortical surface applications and broader 3D mesh domains remains future work. Code is available at https://github.com/labhai/vSV-ViT.
Abstract:Learning from 3D meshes is challenging because the data reside on non-Euclidean surfaces embedded in 3D space. This is particularly evident in domains such as brain cortical surface analysis, where existing models typically rely on ROI-agnostic, face-based, or fixed-size patches. Such patches can duplicate boundary vertices, conflate anatomically distinct regions, or incorporate non-cortical vertices such as those of the medial wall. We develop variable-size supervertex (vSV) partitioning, which groups cortical surface vertices into vSV patches similar to superpixels in images. Building on these vSVs, we design Variable-size SuperVertex Vision Transformer (vSV-ViT), a Vision Transformer that processes vSVs through padding, mask-aware patch embedding, and vertex position embedding. We demonstrate the benefits of the proposed framework through a quantitative analysis of vertex assignment in surface partitioning and downstream tasks using Alzheimer's disease brain image datasets. For downstream tasks, we use meshes of cortical thickness and cortical curvature as input to various encoder models and evaluate classification performance. The superior predictive accuracy of vSV-ViT over recent surface-based models supports its effectiveness for cortical surface learning, with AD-related and sex prediction serving as downstream validation. Extending the framework to other cortical surface applications and broader 3D mesh domains remains future work. Code is available at this https URL.
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
arXiv:2605.26514 [cs.CV]
(or arXiv:2605.26514v2 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2605.26514 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 305 cuvinte
Subiecte conexe
Build a Physical AI model factory with NVIDIA Cosmos 3 on SageMaker HyperPod
AWS · 4 sept. 2026
J++ Lens: Jacobian Filtering Enables More Faithful Workspace Lenses
LessWrong · 8 oct. 2026
Runway’s WorldPrompt and the Engineering of Real-Time Worlds
Latent Space · 25 sept. 2026
The Machines that Make the Machines
NVIDIA · 7 oct. 2026