Surface Vision Transformers: Attention-Based Modelling applied to Cortical Analysis

Simon Dahan, Abdulah Fawaz, Logan Z. J. Williams, Chunhui Yang, Timothy S. Coalson, Matthew F. Glasser, A. David Edwards, Daniel Rueckert, Emma C. Robinson
Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, PMLR 172:282-303, 2022.

Abstract

The extension of convolutional neural networks (CNNs) to non-Euclidean geometries has led to multiple frameworks for studying manifolds. Many of those methods have shown design limitations resulting in poor modelling of long-range associations, as the generalisation of convolutions to irregular surfaces is non-trivial. Motivated by the success of attention-modelling in computer vision, we translate convolution-free vision transformer approaches to surface data, to introduce a domain-agnostic architecture to study any surface data projected onto a spherical manifold. Here, surface patching is achieved by representing spherical data as a sequence of triangular patches, extracted from a subdivided icosphere. A transformer model encodes the sequence of patches via successive multi-head self-attention layers while preserving the sequence resolution. We validate the performance of the proposed Surface Vision Transformer (SiT) on the task of phenotype regression from cortical surface metrics derived from the Developing Human Connectome Project (dHCP). Experiments show that the SiT generally outperforms surface CNNs, while performing comparably on registered and unregistered data. Analysis of transformer attention maps offers strong potential to characterise subtle cognitive developmental patterns.

Cite this Paper

BibTeX
@InProceedings{pmlr-v172-dahan22a, title = {Surface Vision Transformers: Attention-Based Modelling applied to Cortical Analysis}, author = {Dahan, Simon and Fawaz, Abdulah and Williams, Logan Z. J. and Yang, Chunhui and Coalson, Timothy S. and Glasser, Matthew F. and Edwards, A. David and Rueckert, Daniel and Robinson, Emma C.}, booktitle = {Proceedings of The 5th International Conference on Medical Imaging with Deep Learning}, pages = {282--303}, year = {2022}, editor = {Konukoglu, Ender and Menze, Bjoern and Venkataraman, Archana and Baumgartner, Christian and Dou, Qi and Albarqouni, Shadi}, volume = {172}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v172/dahan22a/dahan22a.pdf}, url = {https://proceedings.mlr.press/v172/dahan22a.html}, abstract = {The extension of convolutional neural networks (CNNs) to non-Euclidean geometries has led to multiple frameworks for studying manifolds. Many of those methods have shown design limitations resulting in poor modelling of long-range associations, as the generalisation of convolutions to irregular surfaces is non-trivial. Motivated by the success of attention-modelling in computer vision, we translate convolution-free vision transformer approaches to surface data, to introduce a domain-agnostic architecture to study any surface data projected onto a spherical manifold. Here, surface patching is achieved by representing spherical data as a sequence of triangular patches, extracted from a subdivided icosphere. A transformer model encodes the sequence of patches via successive multi-head self-attention layers while preserving the sequence resolution. We validate the performance of the proposed Surface Vision Transformer (SiT) on the task of phenotype regression from cortical surface metrics derived from the Developing Human Connectome Project (dHCP). Experiments show that the SiT generally outperforms surface CNNs, while performing comparably on registered and unregistered data. Analysis of transformer attention maps offers strong potential to characterise subtle cognitive developmental patterns.} }
Endnote
%0 Conference Paper %T Surface Vision Transformers: Attention-Based Modelling applied to Cortical Analysis %A Simon Dahan %A Abdulah Fawaz %A Logan Z. J. Williams %A Chunhui Yang %A Timothy S. Coalson %A Matthew F. Glasser %A A. David Edwards %A Daniel Rueckert %A Emma C. Robinson %B Proceedings of The 5th International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2022 %E Ender Konukoglu %E Bjoern Menze %E Archana Venkataraman %E Christian Baumgartner %E Qi Dou %E Shadi Albarqouni %F pmlr-v172-dahan22a %I PMLR %P 282--303 %U https://proceedings.mlr.press/v172/dahan22a.html %V 172 %X The extension of convolutional neural networks (CNNs) to non-Euclidean geometries has led to multiple frameworks for studying manifolds. Many of those methods have shown design limitations resulting in poor modelling of long-range associations, as the generalisation of convolutions to irregular surfaces is non-trivial. Motivated by the success of attention-modelling in computer vision, we translate convolution-free vision transformer approaches to surface data, to introduce a domain-agnostic architecture to study any surface data projected onto a spherical manifold. Here, surface patching is achieved by representing spherical data as a sequence of triangular patches, extracted from a subdivided icosphere. A transformer model encodes the sequence of patches via successive multi-head self-attention layers while preserving the sequence resolution. We validate the performance of the proposed Surface Vision Transformer (SiT) on the task of phenotype regression from cortical surface metrics derived from the Developing Human Connectome Project (dHCP). Experiments show that the SiT generally outperforms surface CNNs, while performing comparably on registered and unregistered data. Analysis of transformer attention maps offers strong potential to characterise subtle cognitive developmental patterns.
APA
Dahan, S., Fawaz, A., Williams, L.Z.J., Yang, C., Coalson, T.S., Glasser, M.F., Edwards, A.D., Rueckert, D. & Robinson, E.C.. (2022). Surface Vision Transformers: Attention-Based Modelling applied to Cortical Analysis. Proceedings of The 5th International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 172:282-303 Available from https://proceedings.mlr.press/v172/dahan22a.html.

Related Material