Skip to content

christianpayer/MedicalDataAugmentationTool-HeatmapRegression

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Integrating Spatial Configuration into Heatmap Regression Based CNNs for Landmark Localization

Usage

This example implements the networks of the papers Regressing Heatmaps for Multiple Landmark Localization Using CNNs and Integrating Spatial Configuration into Heatmap Regression Based CNNs for Landmark Localization. Look into the subfolders of this repository for individual examples and more details.

You need to have the MedicalDataAugmentationTool framework downloaded and in you PYTHONPATH for the scripts to work. If you have problems/questions/suggestions about the code, write me a mail!

Train models

Run the main.py files inside the example folders to train the network.

Train and test other datasets

In order to train and test on other datasets, modify the dataset.py files. See the example files and documentation for the specific file formats. Set the parameter save_debug_images = True in order to see, if the network input images are reasonable.

Citation

If you use this code for your research, please cite our MIA paper or MICCAI paper:

@article{Payer2019a,
  title   = {Integrating Spatial Configuration into Heatmap Regression Based {CNNs} for Landmark Localization},
  author  = {Payer, Christian and {\v{S}}tern, Darko and Bischof, Horst and Urschler, Martin},
  journal = {Medical Image Analysis},
  volume  = {54},
  year    = {2019},
  month   = {may},
  pages   = {207--219},
  doi     = {10.1016/j.media.2019.03.007},
}
@inproceedings{Payer2016,
  title     = {Regressing Heatmaps for Multiple Landmark Localization Using {CNNs}},
  author    = {Payer, Christian and {\v{S}}tern, Darko and Bischof, Horst and Urschler, Martin},
  booktitle = {Medical Image Computing and Computer-Assisted Intervention - {MICCAI} 2016},
  doi       = {10.1007/978-3-319-46723-8_27},
  pages     = {230--238},
  year      = {2016},
}

About

No description, website, or topics provided.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages