CoLa-Diff

CoLa-Diff is a Diffusion Model (DM) based framework for multi-modal medical image synthesis. The algorithm is elaborated in our paper CoLa-Diff: Conditional Latent Diffusion Model for Multi-Modal MRI Synthesis.

Overview

CoLa-diff integrates multi-condition from available MRI contrasts in a compact and low-dimensional latent space to guide the generation of missing modality $x\in\mathbb{R}^{H \times W \times 1}$.

BibTeX

@misc{jiang2023coladiff,
      title={CoLa-Diff: Conditional Latent Diffusion Model for Multi-Modal MRI Synthesis}, 
      author={Lan Jiang and Ye Mao and Xi Chen and Xiangfeng Wang and Chao Li},
      year={2023},
      eprint={2303.14081},
      archivePrefix={arXiv},
      primaryClass={eess.IV}
}