AFIDs

An open framework for evaluating correspondence in brain images and teaching neuroanatomy using anatomical fiducial placement

Documentation: https://afidsproject.com

Join our channel:

Introduction

The AFIDs protocol is an anatomical fiducial placement protocol that has been validated and used for teaching at a number of local and global BrainHack related events.
- Human AFIDs protocol
- Macaca AFIDs protocol
The protocol and study framework leverage open resources/tools and has been developed with full transparency in mind so that others may freely use, adopt, and modify.
AFID placement is reproducible, not overtly manually intensive (20-40 minutes once trained), and more sensitive to local registration errors than standard voxel overlap measures.
60+ raters trained to date.

Convert3D tool: https://sourceforge.net/projects/c3d/
- further documentation for c3d: https://github.com/pyushkevich/c3d/blob/master/doc/c3d.md
python requirements (defined in requirements.txt):
- yYAML>=5.3.1
- scikit-learn>=0.23.1
- nibabel>=3.1.0
- pandas>=1.0.5
- numpy>=1.19.4

The input directory with *_T1w.nii.gz and AFIDs .fcsv files should be setup as follows:

input_data/
    ├── sub-<###>_*.fcsv
    ├── sub-<###>_*T1w.nii.gz
    └── ...

.fcsv is the 3D Slicer markups file with all 32 AFIDs defined for the relative T1w image
T1w.nii.gz is the T1w nifti file associated with the AFIDs markups

Edit the config/config.yaml file to include the paths to the following:

Variable	Description
`input_dir`	full path to where the input directory is stored
`output_dir`	full path to where the pipeline should output the data
`c3d_path`	path to the `c3d` install `bin` directory on your system

Clone a copy of this repository to your system:

git clone https://github.com/greydongilmore/autofids-brainhack2020.git

Install the Python dependencies by opening a terminal, changing to the project directory and running:
```
python -m pip install -r requirements.txt
```

All the following commands should be run in the root of the project directory.

Convert source .fcsv files into a 3D mask using c3d:
```
python /workflow/scripts/fcsvToLandmarks.py
```
Running the training pipeline:
```
python /workflow/scripts/c3dTrain.py
```
Running the testing pipeline:
```
python /workflow/scripts/c3dTest.py
```