[6FH5] Pipeline reproduction (SPM - raw)

[bclenet]

Softwares

SPM12b-6906

Input data

raw data

Additional context

see description below

List of tasks

Please tick the boxes below once the corresponding task is finished. 👍

• 👌 A maintainer of the project approved the issue, by assigning a 🏁status: ready for dev label to it.
• 🌳 Create a branch on your fork to start the reproduction.
• 🌅 Create a file team_{team_id}.py inside the narps_open/pipelines/ directory. You can use a file inside narps_open/pipelines/templates as a template if needed.
• 📥 Create a pull request as soon as you completed the previous task.
• 🧠 Write the code for the pipeline, using Nipype and the file architecture described in docs/pipelines.md.
• 📘 Make sure your code is documented enough.
• 🐍 Make sure your code is explicit and conforms with PEP8.
• 🔬 Create tests for your pipeline. You can use files in tests/pipelines/test_team_* as examples.
• 🔬 Make sure your code passes all the tests you created (see docs/testing.md).

---

NARPS team description : 6FH5

General

• teamID : 6FH5
• NV_collection_link : https://neurovault.org/collections/5663/
• results_comments :
  • the whole brain contrast revealed frontalpolar cortex
  • the whole brain contrast revealed activation in bilateral anterior insula, and anterior cingulate cortex.
• preregistered : No
• link_preregistration_form : NA
• regions_definition : Neurosynth (http://neurosynth.org) and its "meta-analyses" tool was used to determine the pre-hypothesized regions. The regions were also checked by the analysis-team's expertise.
• softwares : SPM12b-6906
• general_comments : NA

Exclusions

• n_participants : 107
• exclusions_details : sub-100; this participant's T1 image was not well scanned (distorted and with ghost image), therefore the coregistration with the functional images might be inaccurate.

Preprocessing

• used_fmriprep_data : No
• preprocessing_order : fMRI data preprocessing was performed using SPM12 (Statistical Parametric Mapping; Wellcome Trust Center for Neuroimaging, University College London, London, UK). A voxel displacement map (VDM) was first calculated from the field map to account for the spatial distortion resulting from the magnetic field inhomogeneity (Jezzard and Balaban, 1995; Andersson et al., 2001; Hutton et al., 2002). Incorporating this VDM, the EPI images were then corrected for motion and spatial distortions through realignment and unwarping (Andersson et al., 2001). Image preprocessing continued with slice timing correction using the middle slice of the volume as the reference. The participants’ anatomical images were manually checked and corrected for the origin by resetting it to the AC-PC. The EPI images were then coregistered to this origin-corrected anatomical image. The anatomical image was skull stripped and segmented into gray matter, white matter, and CSF, using the “Segment” tool in SPM12. These gray and white matter images were used in the SPM12 DARTEL toolbox to create individual flow fields as well as a group anatomical template (Ashburner, 2007) The EPI images were then normalized to the MNI space using the respective flow fields through the DARTEL toolbox normalization tool. A Gaussian kernel of 6 mm full-width at half-maximum (FWHM) was used to smooth the EPI images.
• brain_extraction : NA
• segmentation : Segmentation was done with the SPM12b-6906 "Segmentation" module. Tissue probability map was based on the TPM in SPM, as in ../{SPM_dir}/tmp/TMP.nii.
• slice_time_correction : - Software: SPM12b-6906, slice-timing "temporal" module.
  • With respect to motion correction: slice-time correction was done after the realignment.
  • Reference slice: the multi-band slice-time order was provided by the NARPS team; the reference slice was middle one per band -> max(slice_time) / 2.
  • Interpolation type and order:sinc.
• motion_correction : - Software: SPM12b-6906, "realignunwarp" module.
  • Use of non­rigid registration: no.
  • Use of motion susceptibility correction: yes; with the spm fieldmap toolbox.
  • Reference scan: register to the 1st scan per functional run.
  • Image similarity metric: normalized correlation.
  • Interpolation type: trilinear; whether image transformations are combined to allow a single interpolation: yes, -> a mean functional image.
  • Use of any slice­to­volume registration methods, or integrated with slice time correction: no.
• motion :
• gradient_distortion_correction : Fieldmap toolbox, as described above.
• intra_subject_coreg : - Software: SPM12b-6906, "coreg" module.
  • Type of transformation: non-linear, 12 parameters.
  • Cost function: normalized mutual information (nmi).
  • Interpolation method: NA, because normalization was performed laler.
• distortion_correction : Fieldmap toolbox, as described above.
• inter_subject_reg : - Software: SPM12b-6906, normalization with the "Dartel" toolbox.
  • Whether volume and/or surface based registration is used: a template file from the VBM toolbox was used to run Dartel; template file: ../{SPM_dir}/toolbox/vbm8/Template_{1-6}_IXI550_MNI152.nii
  • Image types registered: both T2* and T1
  • Any preprocessing to images: T1, segmentation of gray matter; T2*, single images -> slice-time & motion corrected images, auf*
  • Template space: MNI; modality: T1 + T2*; resolution: both 2x2x2; others see above
  • Additional template transformation: NO
  • Choice of warp: affine transformation used by Dartel, with FWHM [6 6 6].
  • Use of regularization: Dartel Bounding Box -> [-78 -112 -50; 78 76 85].
  • Interpolation type: trilinear
  • Cost function: minimizing the KL divergence between images and the template
  • Cost function mask: none.
• intensity_correction : SPM default.
• intensity_normalization : SPM default: Global (intensity) normalization was not used in 1st-level analyses.
• noise_removal : NA. Only the standard 6 movement parameters were used.
• volume_censoring : NA. Only the standard 6 movement parameters were used.
• spatial_smoothing : - Software: SPM12b-6906, "smooth" module.
  • Smoothing kernal: FWHM [6 6 6] Gaussian smoothing kernel.
  • Filtering: fixed kernel
  • Space: native space (after slice-time correction)
• preprocessing_comments : NA

Analysis

• data_submitted_to_model : - Number of time points: 453 scans per run, 4 runs; except missing trials (see below).
  • Number of subjects: 107; 53 in equal range group, 54 in equal indifference group.
  • Exclusion of time points: non-responses were modeled separately, as a regressor of non-interest.
  • Excusion of subjects: sub-100, because of poor quality T1 image.
• spatial_region_modeled : Whole brain.
• independent_vars_first_level :
  • Event­related design predictors: .
    • Onset regressor: the onset when stimuli (gamble) was presented to the participants.
    • Duration: 4s.
    • Parametric modulators (PM): (1) potential gains, and (2) potential losses; potential losses were orthogonalized with respect to the gains.
  • HRF: Canonical only (no derivatives)
  • Drift regressors: high-pass filter 128s
  • Movement regressors: standard rigid-body parameters (6).
  • Any other nuisance regressors: functional scans where paticipants gave non-responces.
  • Any orthogonalization of regressors: yes, see above. We had two PMs, gains and losses, losses was orthogonalized with respect to gains.
• RT_modeling : none
• movement_modeling : 1
• independent_vars_higher_level :
  • Group effects: equal range group (N=53) and equal indifference group (N=54).
  • Whether or not covariates are split by group: no covariates were used, hence, no group-by-covariate interaction.
  • Other between subject effects: No.
• model_type : Mass univariate.
• model_settings :
  • First-level:
    • Drift model: drift fit with Discrete Cosine Transform basis (128s cut-off)
    • Autocorrelation model: AR(1) in SPM.
  • Second-level:
    • mixed-effects model, OLS estimates.
    • For hypotheses 1-8, one sample t-test was used per group (equal range and equal indifference); for hypothesis 9, two sample (independent sample) t-test was used.
• inference_contrast_effect : 1st-level contrast.
  Contrasts were constructed based on the corresponding parametric modulators per run. For instance, suppose an ideal case, where there were only three regressors per run, one for the onset, one for the win and one for the loss. Because there were 4 runs, the design matrix then had 12 columns (excluding nuisance regressors).
  For the gain PM effect, we had, [0 1 0, 0 1 0, 0 1 0, 0 1 0]; Similarly, for the loss PM effect, we had [0 0 1, 0 0 1, 0 0 1, 0 0 1]. In sum, this configuration tested the overall PM effects of gain and loss, respectively.
• search_region : Whole brain.
• statistic_type :
  • Cluster-wise
  • Cluster-forming threshold: p<0.001 uncorrected, k = 134 (p<.05 FWE-corrected at cluster-level), cluster size was determined by running "CorrClusTh" tool, written by Thomas Nichols, Marko Wilke etc..
  • Neighborhood size: 18-connectivity (SPM default).
• pval_computation : Standard parametric inference.
• multiple_testing_correction : Cluster-wise FWE corrected (random field theory).
• comments_analysis : NA

Categorized for analysis

• region_definition_vmpfc : neurosynth
• region_definition_striatum : neurosynth
• region_definition_amygdala : neurosynth
• analysis_SW : SPM
• analysis_SW_with_version : SPM12
• smoothing_coef : 6
• testing : parametric
• testing_thresh : p<0.001
• correction_method : GRTFWE cluster
• correction_thresh_ : k>134, p<0.05

Derived

• n_participants : 107
• excluded_participants : 100
• func_fwhm : 6
• con_fwhm :

Comments

• excluded_from_narps_analysis : No
• exclusion_comment : Missing much of the central brain.
• reproducibility : 2
• reproducibility_comment :