Bridging Modalities on the Cortex: Surface-based MRI to PET Translation with a Diffusion Bridge
DB-SUiT is a surface-based diffusion bridge generating synthetic FDG-PET from MRI on the cortical manifold, lifting dementia classification 14.2% over MRI.
Researchers introduce DB-SUiT, a surface-based diffusion bridge framework for MRI-to-PET translation operating natively on the folded cortical geometry, built around a conditional Spherical U-shaped vision Transformer (SUiT) with spherical convolutional encoders and bottleneck Transformers. On dementia classification, synthesized PET surfaces improved performance by 14.2% over MRI and 11.3% over volumetric PET, and a blinded reader study achieved 85.5% diagnostic accuracy versus 95.2% for real PET. The model generalized to an external cohort including an unseen dementia subtype without retraining, with code released on GitHub.
- Conditional Spherical U-shaped vision Transformer preserves cortical surface topology
- Synthetic PET surfaces improve dementia classification by 14.2% over MRI
- Blinded reader study: 85.5% accuracy vs 95.2% for real PET
- Generalizes cross-cohort to an unseen dementia subtype without retraining
Full article228 words · extracted from arxiv.org · click to collapse
Cortical hypometabolism measured by Fluorodeoxyglucose Positron Emission Tomography (FDG-PET) is a highly sensitive biomarker for dementia diagnosis. However, high costs, radiation exposure, and limited accessibility constrain its clinical utility. While cross-modal synthesis from Magnetic Resonance Imaging (MRI) offers a promising alternative, existing volumetric generation methods do not explicitly account for the highly folded cortical geometry, where disease-related patterns predominantly reside. To address this, we introduce a novel surface-based diffusion bridge framework DB-SUiT for MRI-to-PET translation that operates natively on the cortical manifold. A conditional Spherical U-shaped vision Transformer (SUiT) is specifically designed to model the intricate cross-modal relationships while preserving surface topology. It combines spherical convolutional encoders for multi-scale surface feature extraction with bottleneck Transformers to capture long-range spatial dependencies, while incorporating demographic and subcortical conditions to refine the synthesis. Evaluated on two datasets, including subjects with different dementia types, DB-SUiT demonstrates high-fidelity synthesis that substantially outperforms other baselines. In automated dementia classification, synthesized PET surfaces improve performance over MRI by 14.2% and PET volumes by 11.3%, approaching the performance of real PET surfaces. In a blinded reader study, synthetic PET achieved 85.5% diagnostic accuracy, compared with 75.8% for MRI and 95.2% for real PET. This further demonstrates cross-cohort and cross-pathology generalization, as the model was evaluated without retraining on an external cohort that included a dementia subtype not represented during training. Our code is available at https://github.com/ai-med/DB-SUiT.
Text extracted automatically; images, tables and formatting may be missing. Original: https://arxiv.org/abs/2609.20147