Skip to content

Stats_Volume

Jump to bottom
Alexandre Routier edited this page Jun 10, 2020 · 3 revisions

statistics-volume - Volume-based mass-univariate analysis with SPM

This pipeline performs statistical analysis (currently group comparison) on volume-based features using the general linear model (GLM) [Friston et al., 1994]. To that aim, the pipeline relies on the tools available in SPM.

Volume-based measurements are analyzed in the IXI549Space (from SPM12). Currently, this pipeline mainly handles gray matter maps obtained from T1 images using the t1-volume pipeline and standardized uptake value ratio (SUVR) maps obtained from FDG PET data using the pet-volume pipeline.

Dependencies

If you only installed the core of Clinica, this pipeline needs the installation of Matlab and SPM, or of SPM standalone, on your computer. You can find how to install these software packages on the third-party page.

Running the pipeline

The pipeline is divided into two sub-pipelines:

  • statistics-volume: performs group comparison but no statistical correction. It generates an SPM report necessary to perform statistical corrections using the second sub-pipeline.
  • statistics-volume-correction: performs family-wise error rate (FWE) or false discovery rate (FDR ) correction at the peak/cluster level. The user has to report the values from the SPM report generated by statistics-volume in order to run this pipeline.

statistics-volume pipeline

The pipeline can be run with the following command line:

clinica run statistics-volume <caps_directory> <subject_visits_with_covariates_tsv> <contrast> <feature_type> <group_id>

where:

  • caps_directory is the output folder containing the results of the t1-volume or pet-volume pipeline and the output of the present command, both in a CAPS hierarchy.
  • subject_visits_with_covariates_tsv is a TSV file containing a list of subjects with their sessions and all the covariates and factors of the model (the content of the file is explained in the Example subsection of the statistics-surface pipeline).
  • contrast is a string defining the contrast matrix or the variable of interest for the GLM, e.g. group or age.
  • feature_type indicates to Clinica which volume-based data to use. It can be either graymatter (outputs of t1-volume) or fdg (outputs of pet-volume). Use 'custom' if you want to use the --custom_files flag (more below).
  • group_id defines the group name for the analysis.

Optional parameters:

  • --group_id_caps is used when you have multiple groups in your CAPS and Clinica is not able to determine which one to choose when reading inputs.
  • -fwhm is the full width at half maximum (FWHM) of the smoothing used in your input file (by default 8 (mm), i.e. the default value of the t1-volume) and pet-volume pipelines)).

statistics-volume-correction pipeline

Once the statistics-volume sub-pipeline has finished, you need to open the SPM report (report1.png or report2.png file). This will look like as follows:

SPM report

You will need to report the following information in the statistics-volume-correction pipeline:

  • height_threshold: T value corresponding to an uncorrected p-value of 0.001
  • FWEp: height threshold (i.e. voxel-level (= peak) threshold)
  • FDRp: height threshold (i.e. voxel-level (= peak) threshold)
  • FWEc: extent threshold (i.e. cluster size threshold)
  • FDRc: extent threshold (i.e. cluster size threshold)

The pipeline can then be run with the following command line:

clinica run statistics-volume-correction  <caps_directory> <t_map> <height_threshold> <FWEp> <FDRp> <FWEc> <FDRc>

where:

  • t_map: name of the T statistic map used for the correction

Optional parameters:

  • n_cuts: number of cuts to display in the final figure

Outputs

Results are stored in the following folder of the CAPS hierarchy: groups/group-<group_label>/statistics_volume/group_comparison_measure-<measure-label>. The most important files are:

  • group-<group_id>_participants.tsv: copy of the subject_visits_with_covariates_tsv parameter file.
  • group-<grp>_<grp_1>-lt-<grp_2>_measure-<msr>_fwhm-<fwhm>_TStatistics.nii: T statistics associated with the hypothesis group1 < group2.
  • group-<grp>_mask.nii: voxels included in the analysis.
  • group-<grp>_report.png: all the results of the 2 sample t-test generated by SPM. Contains information necessary to use the statistics-volume-correction sub-pipeline.

The <group_1>-lt-<group_2> means that the tested hypothesis is: "the measurement of <group_1> is lower than (lt) the measurement of <group_2>". The pipeline includes both contrasts so *<group_2>-lt-<group_1>* files are also saved.

The full list of output files from the statistics-volume-[correction] pipeline can be found in the The ClinicA Processed Structure (CAPS) specifications.

Describing this pipeline in your paper

!!! cite "Example of paragraph:" These results have been obtained using the statistics-volume pipeline of Clinica [Routier et al]. This pipeline is a wrapper of the statistical analysis toolbox implemented in SPM. More precisely, a point-wise, voxel-to-voxel model was used to conduct a group comparison of whole brain voxels. The data were smoothed using a Gaussian kernel with a full width at half maximum (FWHM) set to <FWHM> mm. The general linear model was used to control for the effect of <covariate_1>, ... and <covariate_N>.

- For FWEp: Statistics were corrected for multiple comparisons using the family-wise error (FWE) correction at the peak level with a statistical threshold of P < 0.05 FWE.
- For FWEc: Statistics were corrected for multiple comparisons using the family-wise error (FWE) correction at the cluster level. A statistical threshold of P < `<ClusterThreshold>` was first applied (height threshold). An extent threshold of P < 0.05 corrected for multiple comparisons was then applied at the cluster level.
- For FDRc: Statistics were corrected for multiple comparisons using the false discovery rate (FDR) correction at the cluster level. A statistical threshold of P < `<ClusterThreshold>` was first applied (height threshold). An extent threshold of P < 0.05 corrected for multiple comparisons was then applied at the cluster level.

!!! tip Easily access the papers cited on this page on Zotero.

Support

(Advanced) Using volumes other than gray matter or FDG PET SUVR maps

If you run the help command line clinica run statistics-volume -h, you will find the following flag:

  • --custom_files CUSTOM_FILES: it allows you to specify which file should be taken in the CAPS/subjects directory. For example, if you want to use the file sub-<subject_label>_ses-<session_label>_task-rest_acq-av45_pet_space-Ixi549Space_pet.nii.gz that is contained in CAPS/subjects/sub-<subject_label>/ses-<session_id>/pet/preprocessing/group-<group_label>, you can use the argument *sub-*_ses-*_task-rest_acq-av45_pet_space-Ixi549Space_pet.nii.gz. If you want to specify the group, you can use group-<group_label>/sub-*_ses-*_task-rest_acq-av45_pet_space-Ixi549Space_pet.nii.gz

This flag is read by Clinica when feature_type is neither graymatter nor fdg. The value you set in feature_type will appear in the _measure-<feature_type> key/value of the output files once the pipeline has run.

Clone this wiki locally