3D Shape-to-Image Brownian Bridge Diffusion for Brain MRI Synthesis from Cortical Surfaces

• DOI: 10.48550/arxiv.2502.12742
• Publication Date: 2025-02-18
• Authors: Fabian Bongratz; Yitong Li; Sama Elbaroudy; Christian Wachinger
• Keywords: Bridge (graph theory); Diffusion imaging

Despite recent advances in medical image generation, existing methods struggle to produce anatomically plausible 3D structures. In synthetic brain magnetic resonance images (MRIs), characteristic fissures are often missing, and reconstructed cortical surfaces appear scattered rather than densely convoluted. To address this issue, we introduce Cor2Vox, the first diffusion model-based method that translates continuous shape priors MRIs. achieve this, leverage a Brownian bridge process which allows for direct structured mapping between contours images. Specifically, adapt concept of model extend it embrace various complementary representations. Our experiments demonstrate significant improvements geometric accuracy structures compared previous voxel-based approaches. Moreover, Cor2Vox excels quality diversity, yielding high variation non-target like skull. Finally, highlight capability our approach simulate atrophy at sub-voxel level. code is available https://github.com/ai-med/Cor2Vox.