N

NiftyFit-Release

NiftyFit is an open-source software library to facilitate voxel wise fitting on a number of datatypes including T1 and T2 relaxometry, Arterial Spin Labeled MRI, Diffusion Weighted Imaging and Dynamic Contrast Enhanced MRI.

Name Last Update
CMake Loading commit data...
fit-apps Loading commit data...
fit-io Loading commit data...
fit-lib Loading commit data...
.DS_Store Loading commit data...
CMakeLists.txt Loading commit data...
LICENSE.txt Loading commit data...
README.md Loading commit data...
README_DEVELOPER.txt Loading commit data...
logo_v1.png Loading commit data...
niftyfit_data_v1.zip Loading commit data...
pnglibconf.h Loading commit data...

NiftyFit

NiftyFit is an open-source software library to facilitate voxel wise fitting on a number of datatypes including T1 and T2 relaxometry, Arterial Spin Labeled MRI, Diffusion Weighted Imaging and Dynamic Contrast Enhanced MRI.

NiftyFit was developed at University College London (UCL).

If you use this software, please cite this paper: NiftyFit: A software package for multi-parametric model-fitting of 4D magnetic resonance imaging data. Andrew Melbourne, Nicolas Toussaint, David Owen, Ivor Simpson, Thanasis Anthopoulos, Enrico De Vita, David Atkinson, Sebastien Ourselin. NeuroInformatics 2015. In Press.

License

Copyright (c) 2013, University College London.

NiftyFit is available as free open-source software under a BSD license. Other licenses may apply for dependencies.

Funding

This project was jointly funded by the MRC (MR/J01107X/1), the National Institute for Health Research (NIHR), the EPSRC (EP/H046410/1) and the National Institute for Health Research University College London Hospitals Biomedical Research Centre (NIHR BRC UCLH/UCL High Impact Initiative- BW.mn.BRC10269). This work is supported by the EPSRC-funded UCL Centre for Doctoral Training in Medical Imaging (EP/L016478/1).

Supported Platforms

NiftyFit has been developed on MacOSX, there is no official support for Windows/Linux at this stage.

HOW TO BUILD THE CODE

The code can be easily built using cmake (http://www.cmake.org/). The latest version can be downloaded from http://www.cmake.org/cmake/resources/software.html Assuming that the code source are in the source path folder, you will have to first create a new folder, i.e. build path (step 1) and then to change directory to move into that folder (step 2).

1 >> mkdir build path

2 >> cd build path

There you will need to call ccmake (step 3a) in order to fill in the build options. If you don’t want to specify options, we could just use cmake (step 3b) and the default build values will be used.

3a >> ccmake source path

3b >> cmake source path

Once all the flags are properly filled in, just press the ”c” to configure the Make- file and then the ”g” key to generate them. In the prompt, you just have to make (step 4) first and then make install (step 5).

4 >> make

5 >> make install

Examples

Once installed, the code can be run as follows. Modality specific commands are followed by the 4D input image and a number of options:

e.g:

fit_asl -h fit_dce -h fit_qt2 -h fit_qt1 -h fit_dwi -h

fit_asl -source image.nii (followed by other options) fit_dce -source image.nii (followed by other options) fit_qt2 -source image.nii (followed by other options) fit_qt1 -source image.nii (followed by other options) fit_dwi -source image.nii (followed by other options)

niiinfo - displays basic nifty header information (can use wildcards)

Main ASL variants (PASL):

1) fit_asl -source asl4D.nii -cbf cbf.nii Runs cbf fitting assuming all tissue is GM!

2) fit_asl -source asl4D.nii -calcT1map TI4D.nii -cbf cbf.nii -t1map T1map.nii Runs cbf fitting, but using IR/SR T1 data to estimate the local T1 only. Assumes GM parameters.

3) fit_asl -source asl4D.nii -calcT1map TI4D.nii -cbf cbf.nii -t1map T1map.nii -sig Runs cbf fitting using IR/SR T1 data to estimate the local T1 and uses this data to fit tissue specific blood flow parameters (lambda, transit times).

4) fit_asl -source asl4D.nii -seg seg4D.nii -cbf cbf.nii -T1map t1map.nii Runs cbf fitting using IR/SR T1 data to estimate the local T1 and uses the segmentation data to fit tissue specific blood flow parameters (lambda,transit times,T1).

5) fit_asl -source asl4D.nii -calcT1map TI4D.nii -seg seg4D.nii -cbf cbf.nii -t1map T1map.nii Runs cbf fitting using IR/SR T1 data to estimate the local T1 and uses the segmentation data to fit tissue specific blood flow parameters (lambda, transit times).

6) fit_asl -source asl4D.nii -calcT1map TI4D.nii -seg seg4D.nii -cbf cbf.nii -t1map T1map.nii -pv Runs pv cbf fitting using IR/SR T1 data to estimate the local T1 and uses the segmentation data to fit tissue specific blood flow parameters (lambda, transit times) and generate a vector of [cbf_GM and cbf_WM].

Main qT1 variants:

1) fit_qT1 -source TI4D.nii -T1map t1map.nii -TI 1 2 5 -IR Runs T1 fitting to inversion and saturation recovery data (NLSQR).

2) fit_qT1 -source TI4D.nii -T1map t1map.nii -flips 2 4 8 -SPGR Runs T1 fitting to spoiled gradient echo (SPGR) data (NLSQR).

Main qT2 variants:

1) fit_qT2 -source qT2.nii -t2map T2map.nii -TEs 89 200 Generate a simple single component T2 map (LSQR).

2) fit_qT2 -source qT2.nii -t2map T2map.nii -TE 12 Generate a simple single component T2 map fitted to multi-echo data (NNLSQR).

3) fit_qT2 -source qT2.nii -t2map T2map.nii -TE 12 -nc 5 -comp comp.nii -t2vals vals.nii Generate a multi-component T2 map with nc components fitted to multi-echo data. Outputs are the mean T2 map, component magnitudes and component T2s (NNLSQR).

Main DWI variants:

1) fit_dwi -source dwi.nii -mono -mdmap adc.nii -bval bvals

2) fit_dwi -source dwi.nii -ivim -mdmap adc.nii -mcmap mcmap.nii -bval bvals -bvec bvecs

3) fit_dwi -source dwi.nii -dti -famap fa.nii -bval bvals -bvec bvecs

4) fit_dwi -source dwi.nii -ball -famap fa.nii -mcmap mcmap.nii -bval bvals -bvec bvecs

5) fit_dwi -source dwi.nii -nod -famap fa.nii -mcmap mcmap.nii -bval bvals -bvec bvecs

Main DCE variants:

1) dce -source dce.nii -pmap parameters.nii -aifmask aifmask.nii Fits the Kety model with a cosine-bolus AIF model to every voxel and returns the parameter map for [ve,KT,vp,dt], given an AIF independently fitted to the aifmask ROI (NLSQR).