This file contains the simple instructions for using the
Topographic Tract Filtering (TTF) tool
Junyan Wang
email: junyan.wang@loni.usc.edu
Prerequisites:
1. Matlab
2. Trackvis(http://trackvis.org/) or brainsuite(http://brainsuite.org) for tract visualization
3. Optional (if no Matlab): MATLAB Runtime Library (R2014b) (https://www.mathworks.com/products/compi...).
Main content:
The package contains the conference version of the TTF tooll
tract_filtering_sEV.m written in matlab
1. tract_filtering_sEV.m is the implementation of the work
published in:
Wang, Junyan, Dogu Baran Aydogan, Rohit Varma, Arthur W. Toga, and
Yonggang Shi. "Topographic Regularity for Tract Filtering in Brain
Connectivity." In International Conference on Information
Processing in Medical Imaging (IPMI), pp. 263-274. Springer, Cham,
2017.
We also include precompiled linux executable for this code:
run_tract_filtering_sEV.sh
tract_filtering_sEV
Usage (MATLAB):
1. First-time user may execute demo_tractfilter.m to see some
typical results.
2. Functions:
To use tract_filtering_sEV.m, please follow:
tract_filtering_sEV(input_tckfile,sr,K,th,sigma,numpc,output_tckfile)
input_tckfile: input tract file with extension .trk or .tck
sr: sampling rate for bundle (default 1)
K: size of neighborhood (default 200)
th: threshold for topographic variation (default 0.1)
sigma: parameter for the exponential in the proximity measure
(default 0.01)
numpc: number of minor components to be used (default 1)
output_tckfile: output tract file with extension .trk or .tck
Usage (without MATLAB):
1. Download and install Matlab Runtime Libraries (R2014b) following
the instructions on: https://www.mathworks.com/help/compiler_...
2. First-time user may execute demo_tractfilter.sh to see some
typical results
3. Executables
sh run_tract_filtering_sEV.sh $DIR_MATLAB_Compiler_Runtime
$INPUT_TRK $DOWNSAMPLE_RATE $K $THD $SIGMA $NUMCPS $OUTPUT_TRK
DIR_MATLAB_Compiler_Runtime=Dir of Matlab Compiler Runtime Library
(default /usr/local/MATLAB/MATLAB_Compiler_Runtime/v84)
INPUT_TRK=Complete path of input tract
DOWNSAMPLE_RATE=downsampling rate for bundle (default 1)
K=size of neighborhood (default 200)
THD=threshold for topographic variation (default 0.1)
SIGMA=parameter for the exponential in the proximity measure
(default 0.01)
NUMPCS=number of minor components to be used (default 1)
OUTPUT_TRK=Complete path of output tract
Acknowledgements:
Wang, Junyan, Dogu Baran Aydogan, Rohit Varma, Arthur W. Toga, and
Yonggang Shi. "Topographic Regularity for Tract Filtering in Brain
Connectivity." In International Conference on Information
Processing in Medical Imaging (IPMI), pp. 263-274. Springer, Cham,
2017.