dMRI Visualization with Horizon

This section talks about Horizon workflows in DIPY and how to use them.

To follow this tutorial let’s make sure we have the latest version of dipy and fury installed on our system.

pip install dipy --upgrade
pip install fury --upgrade

Let’s explore the options that horizon provides.

dipy_horizon --help

Visualize 3D Brain Image

This tutorial provides a basic example of loading an dMRI image to horizon using the command line interface.

Using a terminal, let’s download a dataset called mni_template. You can skip this step if you already have the dataset downloaded.

dipy_fetch mni_template

To see more details about dipy_fetch you can refer to Downloading DIPY datasets

This command will download the data in your .dipy folder placed in your home directory.

Let’s try to load the image.

For macOS and Linux

dipy_horizon ~/.dipy/mni_template/mni_icbm152_t1_tal_nlin_asym_09a.nii ~/.dipy/mni_template/mni_icbm152_t2_tal_nlin_asym_09a.nii

For Windows(cmd)

dipy_horizon %USERPROFILE%/.dipy/mni_template/mni_icbm152_t1_tal_nlin_asym_09a.nii %USERPROFILE%/.dipy/mni_template/mni_icbm152_t2_tal_nlin_asym_09a.nii

For Windows(powershell)

dipy_horizon "$home/.dipy/mni_template/mni_icbm152_t1_tal_nlin_asym_09a.nii" "$home/.dipy/mni_template/mni_icbm152_t2_tal_nlin_asym_09a.nii"

You can also use your own data by invoking the command below.

dipy_horizon <your_file_name>.nii

We also support direct visualization of compressed NIFTI files with extension .nii.gz.

dipy_horizon <your_fist_file_name>.nii.gz

Visualize 4D Brain Image

This tutorial shows how visualize a 3D image with Volume.

Using a terminal, let’s download a dataset called stanford_hardi. You can skip this step if you already have the dataset downloaded.

dipy_fetch stanford_hardi

This dataset has a NIFTI file with volumes.

Let’s try to load the image.

For macOS and Linux

dipy_horizon ~/.dipy/stanford_hardi/HARDI150.nii.gz

For Windows(cmd)

dipy_horizon %USERPROFILE%/.dipy/stanford_hardi/HARDI150.nii.gz

For Windows(powershell)

dipy_horizon "$home/.dipy/stanford_hardi/HARDI150.nii.gz"

Visualize Brain Tractogram

This tutorial shows how to visualize a tractogram.

Using a terminal, let’s download a dataset called bundle_atlas_hcp842. You can skip this step if you already have the dataset downloaded.

dipy_fetch bundle_atlas_hcp842

Horizon supports below mentioned tractogram formats.

• .trk

• .trx

• .dpy

• .tck

• .vtk

• .vtp

• .fib

Let’s try to load the tractogram.

For macOS and Linux

dipy_horizon ~/.dipy/bundle_atlas_hcp842/Atlas_80_Bundles/whole_brain/whole_brain_MNI.trk --cluster

For Windows(cmd)

dipy_horizon %USERPROFILE%/.dipy/bundle_atlas_hcp842/Atlas_80_Bundles/whole_brain/whole_brain_MNI.trk --cluster

For Windows(powershell)

dipy_horizon "$home/.dipy/bundle_atlas_hcp842/Atlas_80_Bundles/whole_brain/whole_brain_MNI.trk" --cluster

Using the --cluster option, we visualize the clusters(bundles) of the tractograms. If we do not provide --cluster it will open up all the streamlines and the interaction panel will not be provided. Opening the streamlines can be computationally expensive, if a large dataset is provided.

Visualize Brain Surface

This tutorial shows how to visualize surfaces in the Horizon.

Using terminal, let’s download brain surface.

For macOS and Linux

wget https://github.com/maharshi-gor/dipy_data/raw/surface_data/surfaces/lh.pial

For Windows(powershell)

wget https://github.com/maharshi-gor/dipy_data/raw/surface_data/surfaces/lh.pial -O lh.pial

For macOS users, if you do not have wget on your terminal you can setup by writing following command

brew install wget

If you are still getting an error you can download the surface by clicking here.

Previous step will download the file into your current directory.

Downloaded using wget To load the surface,

dipy_horizon lh.pial

Downloaded using link To load the surface,

dipy_horizon <PATH_TO_YOUR_DIRECTORY>/lh.pial