app-classifyber-segmentation

This App implements only the TEST PHASE of Classifyber, i.e. it runs Classifyber as a pre-trained bundle segmentation method. Classifyber is a supervised streamline-based method that performs automatic bundle segmentation based on binary linear classification, which simultaneously combines information from bundle geometries, connectivity patterns, and atlases. In addition, it is robust to a multitude of diverse settings, i.e. it can deal with different bundle sizes, tracking algorithms, and dMRI data qualities.

If you want to run both the TRAINING and TEST PHASES, please refer to the App https://doi.org/10.25663/brainlife.app.228.

Authors

• Giulia Bertò (giulia.berto.4@gmail.com)

Contributors

• Emanuele Olivetti (olivetti@fbk.eu)

Funding Acknowledgement

We kindly ask that you acknowledge the funding below in your publications and code reusing this code.

Citation

We kindly ask that you cite the following article when publishing papers and code using this code:
"Classifyber, a robust streamline-based linear classifier for white matter bundle segmentation", Bertò, G., Bullock, D., Astolfi, P., Hayashi, S., Zigiotto, L., Annicchiarico, L., Corsini, F., De Benedictis, A., Sarubbo, S., Pestilli, F., Avesani, P., Olivetti, E. NeuroImage (2020).

Running the app

On Brainlife.io

You can submit this App online at https://doi.org/10.25663/brainlife.app.265 via the “Execute” tab.

Branch 1.0:

Inputs:
You only need to provide the tractogram of the (target) subject you want to extract the bundle(s) from and the anatomical T1 of the (target) subject. WARNING: the tractogram needs to be already co-registered in the MNI152 T1 space (you can use the App https://doi.org/10.25663/brainlife.app.202 to warp your .tck file to the correct anatomical space).

Output:
You will get the wmc segmentation of the bundle(s) of interest in the target subject. You can convert it in multiple .tck files with the App https://doi.org/10.25663/brainlife.app.251.

You can choose the bundle(s) to be segmented by providing the id(s) related to the training results you want to use as follows:

HCP-IFOF (WARNING: it might be uneffective if the input tractogram has not ~500K streamlines):
32 - Left IFOF
33 - Right IFOF

HCP-minor (WARNING: it might be uneffective if the input tractogram has not ~500K streamlines):
38 - Left pArc
39 - Right pArc
40 - Left TP-SPL
41 - Right TP-SPL
42 - Left MdLF-SPL
43 - Right MdLF-SPL
44 - Left MdLF-Ang
45 - Right MdLF-Ang

Branch 1.1:

Inputs:
You only need to provide the tractogram of the (target) subject you want to extract the bundle(s) from and the anatomical T1 of the (target) subject (the two inputs should be in the same anatomical space). The tractogram will be then internally co-registered to the MNI152 T1 space.

Additional options with respect to branch 1.0:

• initial non-linear registration of the tractogram in the proper MNI152 T1 space
• the output segmented bundles are given also in tck format
• possibility to segment also bundles conforming those of the HCP-major and Clinical datasets as follows:

HCP-major (WARNING: it might be uneffective if the input tractogram has <5M streamlines):
21 - Left CST
22 - Right CST
23 - Left IFOF
24 - Right IFOF
25 - Left ILF
26 - Right ILF
27 - Left UF
28 - Right UF
29 - Left AF
30 - Right AF

Clinical: (WARNING: it might be uneffective if the input tractogram has >300K streamlines)
35 - Left IFOF
36 - Left AF

Branch 1.2:

As branch 1.1 with some small enhancements.

Branch 1.3:

As branch 1.2, but running with python3.

Running locally

1. git clone this repo.
2. Inside the cloned directory, create config.json with something like the following content with paths to your input files:

{
  "tractogram_static": "./track.tck",
  "t1_static": "./t1.nii.gz",
  "tractID_list": "32, 33, 40, 45"
}

3. Launch the App by executing main.

./main

Dependencies

This App only requires singularity to run.

MIT Copyright (c) 2020 Bruno Kessler Foundation (FBK)