GeezerVison

We propose an effective spherical deformation-based sulcal labeling method that automatically enlarges sulcal ROIs while contracting unlabeled regions. By designing an invertible, learnable isotropic deformation field with end-to-end SO(3)-equivariance, the proposed method enhances expressive power and generalization capability under data scarcity.

[WACV 2027] AdaPhase: Navigating the Trajectory with Adaptive Frequency Modulation for Content-preserving Stylization

Rotation-invariant multiscale modeling with spherical harmonics and Mamba for cortical surface analysis and brain age prediction.

SPHARM-Net: Spherical Harmonics-based Convolutional Neural Network

MICCAI 2026 Shape-Adaptive Guidance Signal for Interactive Cortical Sulcal Labeling

MICCAI 2026 Sulcus-Aware Hippocampal Surface Modeling with Training-time Sulcus-guided Learning

Leveraging Input-Level Feature Deformation with Guided-Attention for Sulcal Labeling