KULeuven Neuro Imaging Suite (KUL_NIS)

KUL_NIS provides tools:

• for conversion of dicom data to BIDS format of

  • T1w (with and without Gd)
  • T2w
  • FLAIR
  • PDw
  • FGATIR
  • DIR (double inversion recovery, in development)
  • ASL (not fully BIDS compatible yet)
  • MTR
  • fMRI
  • dMRI
  • QSM (in development)

• for analysis of

  • fMRI data

    • using fmriprep
    • resting stat analysis using FSL melodic
    • task based fMRI

  • diffusion MRI data

    • preprocessing using mrtrix3
    • coregistration to T1w data using synb0-disco and/or Ants
    • normalisation to MNI space using fmriprep
    • group fixel based analysis in group template space

  • myelin markers such as:

    • MRT (magnetisation transfer ratio)
    • T1/T2 and T1/FLAIR ratio's

  • structural data

    • coregistration of data to the T1w without Gd
    • Freesufer parcellation
    • FastSurfer parcellation
    • Tumor segmentation

• to work easily with KUL_VBG

  • a pipeline that enables to run Freesurfer and FastSurfer in patients with (large) brain lesions (tumour/stroke)

• to work easily with KUL_FWT

  • an automated csd probabilistic tractography pipeline

• for conversion of the output of pipelines back to DICOM for use in:

  • Brainlab Neurosurgery format using Karawun
  • conversion of results for import into a general pacs using a MevisLab interface

All scripts (only) work with BIDS data.

Note that some output data of the analysis pipelines can be converted back to dicom for import into a PACS system (see above).

The pipelines used should work fine with healthy volunteer data, but are being implemented for use with clinical data (tumors, stroke, MS, PD) but have not yet been tested fully with clinical data.

Any use in a clinical environment is off-label, not FDA aproved, not CE-labeled or approved. Also see the license file please.

Tools for BIDS data conversion

KUL_dcm2bids

Converts dicom data to BIDS format for multipule MRI vendors. Provides slice-timing, total-readout-time, phase encoding direction & other data for Philips scanners (which is automatically defined for Siemens and GE scanners). Click the link in the header above for more information.

KUL_multisubjects_dcm2bids

KUL_dcm2bids to convert multiple datasets at once.

KUL_bids_summary

Provides output of multiple parameters of a BIDS dataset, including acquisition date, scanner software verion, etc... readable in google sheets, excel, etc...

Tools for structural analysis

KUL_anat_biascorrect

This script performs Ants N4 bias correction on the structural images of a participant in the BIDS folder. Run KUL_anat_biascorrect.sh to see information.

KUL_anat_register_rigid

This script rigidly registers structural images in the BIDS folder to the T1w (without contrast) in the BIDS folder.

KUL_anat_segment_tumor

This script segments pre- or post-operative brain tumor lesions and/or resection cavities using a combination of AI tools hd-glio-auto, resseg and fastsurfer.

Tools for fMRI analysis

KUL_preproc_all

This script allows to start an fmriprep analysis with a config file. See

KUL_fmriproc_spm

This script will run an automated analysis using SPM12 of a standard blocked design active fMRI data with a paradigm using 30 seconds BASELINE followed by 30 seconds TASK epochs, after these have been preprocessed with fmriprep.

KUL_fmriproc_conn

This script will run an automated analysis using FSL melodic on active and resting-state fMRI data, after these have been preprocessed with fmriprep.

Tools for dMRI analysis

KUL_dwiprep

KUL_dwiprep_anat

KUL_dwiprep_MNI

Tools for the study of myelin

Magnetisation Transfer Ratio and T1/T2 and T1/FLAIR ratio as a myelin marker

KUL_T1T2FLAIRMTR_ratio

Tools for importing results back into dicom and transfer to PACS/BrainLab

Karawun

MevisLab

Other (under dev)

Who are we

Dr. Ahmed Radwan - KUL - [email protected]

Prof. Dr. Stefan Sunaert - UZ/KUL - [email protected]