Surface Vision Transformers: Attention-Based Modelling applied to Cortical Analysis
This work addresses limitations in cortical analysis for neuroscience by providing a domain-agnostic method, though it is incremental as it adapts existing vision transformer approaches to surface data.
The authors tackled the problem of modeling long-range associations on non-Euclidean surfaces by introducing a Surface Vision Transformer (SiT) that applies attention-based modeling to cortical data, achieving performance that generally outperforms surface CNNs on phenotype regression tasks.
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.