Preprocessing of MRI for better use in medical tools or AI activities.
In the image below only the axial axis result is shown, but the pipeline works on 3D images.
The following software must be installed to use Fine-Prepro:
The file requirements.txt contains the libraries used for operation.
The files must be .nii.gz.
usage: fine-prepro.py [-h] --inputs INPUTS --outputs OUTPUTS [--threads THREADS] [--norm NORM]
[--robustfov ROBUSTFOV] [--brain_extraction BRAIN_EXTRACTION]
[--lperc LPERC] [--uperc UPERC]
*****{Fine-Prepro}*****
options:
-h, --help show this help message and exit
--inputs INPUTS The input can be a single path or a folder containing T1 MRIs.
--outputs OUTPUTS The output it must be a path for a folder, if folder don't exist one will be created.
--threads THREADS Threads (default: number of cores).
--norm NORM Remove outlier intensities from a brain component, similar to Tukey's fences method.
--robustfov ROBUSTFOV
This flag hallows use of robustfov, whit center the brain, eliminating slices from the z axis (default use robustfov).
--brain_extraction BRAIN_EXTRACTION
If this parameter is setup to 1 the process will produce the brain extracion (default = 1 yes).
Normalizzation optzions:
--lperc LPERC Indicate the lower percentile number you want to consider (default 0).
--uperc UPERC Indicate the upper percentile number you want to consider (default 99).
The input (--inputs) can be:
- a path to nii.gz image
./path/to/.nii.gz
- a path to folder that contains MRIs
./path/to/folder
The outuput (--outputs) must be a path to a folder, if it does't exist it will be created.
./path/to/output_folder
| Phase | Algortihm | Library |
|---|---|---|
| Registration | registration | ANTsPy |
| Normalization | python |
|
| ROI | robustfov | FSL |
| Skull-stripping | mri_synthstrip | ANTs |
| Brain-skull | python |
Where it says python means that code was used to perform the task