icon-lab / SelfRDB

Official PyTorch implementation of SelfRDB, a diffusion bridge model for multi-modal medical image synthesis.
MIT License
10 stars 0 forks source link

about the datasets #2

Open tthjjg opened 3 weeks ago

tthjjg commented 3 weeks ago

This work, like the previous Syndiff paper, is quite impressive and has inspired me a lot. I would like to ask how to handle the MRI-CT dataset, especially since CT images have artifacts. Did the authors select 90 slices per patient? I noticed that some slices have poor quality.

tthjjg commented 3 weeks ago

CT images have white streaks.