Merge pull request #8 from uncbiag/add_registration_examples

Add registration examples

README.md

@@ -5,7 +5,7 @@
 This the official repository for `uniGradICON`: A Foundation Model for Medical Image Registration
 
 `uniGradICON` is based on [GradICON](https://github.com/uncbiag/ICON) but trained on several different datasets (see details below). 
-The result is a deep-learning-based registration model that works well across datasets. 
+The result is a deep-learning-based registration model that works well across datasets. More results can be found [here](/demos/Examples.md).
 
 ![teaser](IntroFigure.jpg?raw=true)
 

demos/Examples.md

@@ -0,0 +1,24 @@
+# Qualitative Results of uniGradICON
+
+## ACDC Dataset
+We evaluated the performance of uniGradICON on the [ACDC dataset](https://www.creatis.insa-lyon.fr/Challenge/acdc/) <sup>[[1]](#1)</sup> by registering the MRIs from the first time point to all subsequent time points. As an example, we've included the target images and the corresponding warped images below. 
+
+![ACDC_target](/demos/Figures/ACDC_target.gif) ![ACDC_warped](/demos/Figures/ACDC_warped.gif)
+
+## DirLab 4DCT Dataset
+We evaluated the performance of uniGradICON on the [4DCT dataset](https://med.emory.edu/departments/radiation-oncology/research-laboratories/deformable-image-registration/index.html) <sup>[[2]](#2)</sup> by registering the CTs from the first time point to all subsequent time points. As an example, we've included the target images and the corresponding warped images below. 
+
+You can acieve
+
+![4DCT_target](/demos/Figures/4DCT_target.gif) ![4DCT_warped](/demos/Figures/4DCT_warped.gif)
+
+## Examples in the paper
+Here is the qualitative results we presented in the [paper](https://arxiv.org/abs/2403.05780).
+
+![Examples_in_paper](/demos/Figures/Registration_examples.png)
+
+## References
+
+<a id="1">[1]</a> Bernard, Olivier, et al. "Deep learning techniques for automatic MRI cardiac multi-structures segmentation and diagnosis: is the problem solved?." IEEE transactions on medical imaging 37.11 (2018): 2514-2525. doi: 10.1109/TMI.2018.2837502
+
+<a id="2">[2]</a> Castillo, Richard, et al. "A framework for evaluation of deformable image registration spatial accuracy using large landmark point sets." Physics in Medicine & Biology 54.7 (2009): 1849.