This uses the FiNE MATLAB tool (release pending) to construct structural networks from a tractogram and a labeled anatomical volume. It will output multiple adjacency matrices based on user requested inputs. It will also produce edge-wise profiles of shape and microstructure properties (also based on user request). The initial version of a LiFE-less network tool.
This will be extended output .json-graph data and other types of networks as the projects around FiNE are finished and published.
The inputs are a processessed tractogram and a parcellation object.
The outputs are a series of network adjacency matrices (will be extended to .json-graph and other network file types.
brainlife.io is publicly funded and for the sustainability of the project it is helpful to Acknowledge the use of the platform. We kindly ask that you acknowledge the funding below in your publications and code reusing this code.
We kindly ask that you cite the following articles when publishing papers and code using this code.
- Avesani, P., McPherson, B., Hayashi, S. et al. The open diffusion data derivatives, brain data upcycling via integrated publishing of derivatives and reproducible open cloud services. Sci Data 6, 69 (2019). https://doi.org/10.1038/s41597-019-0073-y
MIT Copyright (c) 2020 Brent McPherson, brainlife.io, Indiana University, and The University of Texas at Austin
You can submit this App online at https://doi.org/10.25663/bl.app.1 via the "Execute" tab.
- git clone this repo.
- Inside the cloned directory, create
config.jsonwith something like the following content with paths to your input files.
{
"parc": "./input/parc.nii.gz",
"track": "./input/track.tck",
"tensor": "./input/tensor/",
"mask": "./input/mask.nii.gz"
}- Launch the App by executing
main
./mainIf you don't have your own input file, you can download sample datasets from Brainlife.io, or you can use Brainlife CLI.
npm install -g brainlife
bl login
mkdir input
bl dataset download 5a0e604116e499548135de87 && mv 5a0e604116e499548135de87 input/parc
bl dataset download 5a0dcb1216e499548135dd27 && mv 5a0dcb1216e499548135dd27 input/track
All output files will be generated under the current working directory (pwd). The main output of this App is a folder called output. This file contains following object.
output:
count.csv - adjacency matrix file of streamline counts
density.csv - adjacency matrix file of streamline density
length.csv - adjacency matrix file of streamline length
denlen.csv - adjacency matrix file of streamline density corrected for length
*_mean.csv - adjacency matrix file of average microstructure property
*_std.csv - adjacency matrix file of standard deviation of microstructure property
stats.json - a series of network statistics estimated on the networks without thresholding.
*_tp.csv - a list of tract profiles (rows) of requested nodes (columns) of the requested microstructure for every edge that exists.
curv.csv - a list of tract profiles (rows) of requested nodes (columns) of the tract curvature for every edge that exists.
tors.csv - a list of tract profiles (rows) of requested nodes (columns) of the tract torsion for every edge that exists.
centers.csv - a list of node centers in real world (X,Y,Z) mm coordinates
pconn.mat - MATLAB object containing connection (edge) data - for debugging / advanced offline work
rois.mat - MATLAB object containing ROI (node) data - for debugging / advanced offline work
omat.mat - MATLAB object containing matrix data - for debugging / advanced offline wor
olab.mat - MATLAB object containing label data - for debugging / advanced offline work
The secondary output of this app is product.json. This file allows web interfaces, DB and API calls on the results of the processing.
This App only requires singularity to run. If you don't have singularity, you will need to install following dependencies.