ImgX-DiffSeg: 3D Medical Image Segmentation Based on DDPMs
Table of contents
- foreword
- currently existing problems
- ImgX-DiffSeg Architecture
- overview
- DDPM with Variance Schedule Resampling
- Diffusion Model for Segmentation
- experiment
- Summarize
- reference
This article was first published on GiantPandaCV, and may not be reproduced without permission! !
foreword
This article continues to interpret the medical image diffusion series. We have previously introduced the application of diffusion in self-supervised and supervised segmentation. Link:
-
https://mp.weixin.qq.com/s/7g3_4hHfKCAp2WQibkbzgA
-
https://mp.weixin.qq.com/s/TR6TaTGAdzQZNPUgSRARrQ
This article "Importance of Aligning Training Strategy with Evaluation for Diffusion Models in 3D Multiclass Segmentation" is not a new diffusion application, but optimizes the training and reasoning strategies and adapts to 3D medical image segmentation tasks , the reference link is at the end of the article.
currently existing problems
- Currently, architecture training and reasoning with a diffusion model are time-consuming.
- In some split tasks, the diffusion mod is not sure