This example shows how to extract brain information and mask from an input b0 image using dipy, a diffusion imaging analysis package. This is based on an example available at dipy's gallery.
The original experiment is composed by a single script named brain-segmentation.py, which consumes and segments the brain data that comes from a 1.5 tesla Siemens MRI (siemens_scil_b0).
To run this experiment without ReproZip, you will need to first install dipy.
The ReproZip package is available here (46.5 MB).
The experiment can be reproduced as follows:
$ reprounzip vagrant setup brain-segmentation.rpz brain-segmentation/
$ reprounzip vagrant run --enable-x11 brain-segmentation/
The output image with the results can be retrieved as follows:
$ reprounzip vagrant download brain-segmentation/ median_otsu.png
You can also perform the same prediction with an alternate input data (ge_scil_b0) as follows:
$ reprounzip vagrant upload brain-segmentation/ ge_scil_b0:siemens_scil_b0
$ reprounzip vagrant run --enable-x11 brain-segmentation/
And finally download the results as follows:
$ reprounzip vagrant download brain-segmentation/ median_otsu.png
If you are using our demo VM image, you can run the following:
$ vagrant ssh
$ cd reprozip-examples/brain-segmentation/
$ python brain-segmentation.py