jsheunis / multi-echo-super Goto Github PK

I'm picking this effort up again after receiving a jolt of motivation during the multi-echo community meeting at OHBM 2022. The idea is to curate a datalad "super"dataset that has subdatasets for all known and publicly available multi-echo datasets out there (in any format: BIDS / non-BIDS / raw / derivative). This can then be used as the basis for metadata extraction and catalog generation using datalad-catalog, which means we can easily host a user friendly browser-based catalog of multi-echo fMRI data! Since everything is structured into a datalad dataset, it also means that data access will be streamlined.

I've started making a list here (results from a search on OpenNeuro and from the list of datasets on tedana):

• Multi-echo fMRI replication sample of autobiographical memory, prospection and theory of mind reasoning tasks
  • OpenNeuroDatasets url: https://github.com/OpenNeuroDatasets/ds000210
  • Path in superdataset: raw/autobiographical_memory
• Multi-echo Cambridge
  • OpenNeuroDatasets url: https://github.com/OpenNeuroDatasets/ds000258
  • Path of raw data in superdataset: raw/multi_echo_cambridge
  • Derivative dataset (fmriprep): https://github.com/jsheunis/multi-echo-cambridge-fmriprep
  • Path of dervative data in superdataset: derivatives/multi_echo_cambridge_fmriprep
• Multiband multi-echo imaging of simultaneous oxygenation and flow timeseries for resting state connectivity
  • OpenNeuroDatasets url: https://github.com/OpenNeuroDatasets/ds000216
  • Includes mriqc derivatives
  • Path in superdataset: raw/simultaneous_asl_bold_resting
• Multiband Multi-Echo Simultaneous ASL/BOLD for task-induced functional MRI
  • OpenNeuroDatasets url: https://github.com/OpenNeuroDatasets/ds000254
  • Path in superdataset: raw/simultaneous_asl_bold_task
• Valence processing differs across stimulus modalities
  • OpenNeuroDatasets url: https://github.com/OpenNeuroDatasets/ds001491
  • Path in superdataset: raw/valence_processing
• Multi-echo masking test dataset
  • OpenNeuroDatasets url: https://github.com/OpenNeuroDatasets/ds002156
  • Path in superdataset: raw/masking_test
  • Derivative dataset (fmriprep): https://github.com/jsheunis/masking-test-fmriprep
  • Path of dervative data in superdataset: derivatives/masking_test_fmriprep
• Le Petit Prince: A multilingual fMRI corpus using ecological stimuli
  • OpenNeuroDatasets url: https://github.com/OpenNeuroDatasets/ds003643
  • Path in superdataset: raw/le_petit_prince
  • Derivative dataset (fmriprep): progress being tracked in this issue
  • TODO: create subdataset of derivatives once available, add to superdataset
• Cambridge Centre for Ageing Neuroscience (Cam-CAN)
  • URL: https://camcan-archive.mrc-cbu.cam.ac.uk/dataaccess/
  • Data access request required
  • TODO: create minimalistic datalad dataset with dataset-level metadata that links user to access request process.
• rt-me-fMRI - A task and resting state dataset for real-time, multi-echo fMRI methods development and validation
  • URL: https://doi.org/10.34894/R1TNL8
  • Data access request required
  • TODO: create minimalistic datalad dataset with dataset-level metadata that links user to access request process.
  • TODO: check if datalad-dataverse extension can be used to mirror the dataset as a datalad dataset
• EuskalIBUR
  • OpenNeuroDatasets url: https://github.com/OpenNeuroDatasets/ds003192
  • TODO: add as subdataset to superdataset
• Neurocognitive aging data release with behavioral, structural, and multi-echo functional MRI measures
  • OpenNeuroDatasets url: https://github.com/OpenNeuroDatasets/ds003592
  • TODO: figure out why the github repo is empty, and try to (help) resolve this so that the datalad dataset can be used (comment left on OpenNeuro dataset: https://openneuro.org/datasets/ds003592/versions/1.0.11)
  • TODO: add as subdataset to superdataset
• Heart rate variability biofeedback training and emotion regulation
  • OpenNeuroDatasets url: https://github.com/OpenNeuroDatasets/ds003823
  • TODO: add as subdataset to superdataset
  • TODO: github repo is not a correct datalad dataset (although already installed as subdataset to multi-echo-super) ==> fix this (either on OpenNeuro side, ideally, or with an interim step)

Please add whatever is missing. Tagging @emdupre @smoia @dowdlelt @jbteves @tsalo @eurunuela @handwerkerd

• ME Cambridge fmriprep derivatives: https://osf.io/9wcb8/ (see https://github.com/jsheunis/multi-echo-cambridge-fmriprep)
• Masking: https://osf.io/4d8cx/
• Le Petit Prince: follow progress here: ME-ICA/tedana#785

For example, the T1w images in ds000210 are 4D and need to be reduced to 3D before fMRIPrep can be run on them. Should the modified dataset be published to G-Node GIN and then added to the super-dataset from there instead of using the one directly from OpenNeuro?