Skip to content

Latest commit

 

History

History
133 lines (105 loc) · 5.75 KB

README.md

File metadata and controls

133 lines (105 loc) · 5.75 KB
Raw

bidsphysio

Converts physio data (CMRR, AcqKnowledge, Siemens PMU) to BIDS physiological recording

Docker image TravisCI CodeCoverage DOI

Usage - Single File

You can use physio2bidsphysio to convert a single file:

physio2bidsphysio --infile <physiofiles> --bidsprefix  <Prefix> [--verbose]

Example:

physio2bidsphysio --infile 07+epi_MB4_2mm_Physio+00001.dcm      \
                  --bidsprefix BIDSfolder/sub-01/func/sub-01_task-REST_run-1

Arguments

  • <physiofiles> space-separated files with the physiological recordings.

    Supported file types:

    • <.dcm> or <.log>: DICOM or log file with physiological recording from a CMRR Multi-Band sequence (only a single file for <.dcm>).
    • <.acq>: AcqKnowledge file (BioPac).
    • <.puls> or <.resp>: Siemens PMU file (VB15A, VBX, VE11C).
    • <.edf> SR Research .edf file

  • <Prefix> is the prefix that will be used for the BIDS physiology files. If all physiological recordings have the same sampling rate and starting time, the script will save the files: <Prefix>_physio.json and <Prefix>_physio.tsv.gz. If the physiological signals have different sampling rates and/or starting times, the script will save the files: <Prefix>_recording-<label>_physio.json and <Prefix>_recording-<label>_physio.tsv.gz, with the corresponding labels (e.g., cardiac, respiratory, etc.).

  • --verbose will print out some warning messages.

Note: If desired, you can use the corresponding _bold.nii.gz BIDS file as --bidsprefix. The script will strip the _bold.nii.gz part from the filename and use what is left as <Prefix>. This way, you can assure that the output physiology files match the _bold.nii.gz file for which they are intended.

Usage - Full session

Alternatively, you can convert a whole session worth of physiology files automatically naming them. Currently, AcqKnowledge and eyetracking .edf files are supported:

acqsession2bids --infolder <physiofolder> --bidsfolder <bidsfolder> --subject <subID>

edfsession2bids --infolder <physiofolder> --bidsfolder <bidsfolder> --subject <subID>

The tool will find which physiological file corresponds to which functional image file and will name it accordingly.

Arguments

  • <physiofolder>: Path to a folder containing all the .acq files for a full session.
  • <bidsfolder>: Path to the top level BIDS folder where you want to extract the physiological data. The functional images corresponding to this session need to have been extracted.
  • <subID>: label of the participant to whom the physiological data belong. The label corresponds to sub-<participant_label> from the BIDS spec (so it does not include "sub-").

Installation

You can install the tool directly from PyPI:

pip install bidsphysio

If you don't want to install the whole package, you can install individual subpackages:

pip install bidsphysio.<sub-package>

Available sub-packages are acq2bids (for .acq files), dcm2bids (for .dcm and .log CMRR physiology files) , pmu2bids (for Siemens PMU files) and edf2bids (for SR Research eyetracking .edf files). You can also install the base classes with the bidsphysio.base sub-package.

Alternatively, you can download the package and install it with pip:

mkdir /tmp/bidsphysio && \
    curl -sSL https://github.com/cbinyu/bidsphysio/archive/master.tar.gz \
        | tar -vxz -C /tmp/bidsphysio --strip-components=1 && \
    cd /tmp/bidsphysio && \
    pip install . && \
    cd / && \
    rm -rf /tmp/bidsphysio

Docker usage

After pulling the latest version (docker pull cbinyu/bidsphysio), start a Docker container with:

docker run [docker options] -v <datapath_host>:/data -v <outpath_host>:/output --entrypoint=/opt/venv/bin/python cbinyu/bidsphysio /opt/venv/bin/edf2bidsphysio --infile /data/<filename>.edf --bidsprefix /output/sub-01

For example, using the test dataset in this repository:

docker run --rm -it -v $PWD/bidsphysio.edf2bids/tests/data/:/data -v $PWD/output:/output --entrypoint=/opt/venv/bin/python cbinyu/bidsphysio /opt/venv/bin/edf2bidsphysio --infile /data/sample.edf --bidsprefix /output/sub-01/func/sub-01

Then, you can run any bidsphysio command (see Usage above)

Alternative use:

Installing bidsphysio will also install the binaries: dcm2bidsphysio, acq2bidsphysio and pmu2bidsphysio, to extract a specific file type:

dcm2bidsphysio --infile <DCMfile> --bidsprefix <Prefix>
acq2bidsphysio --infile <acqfiles> --bidsprefix <Prefix>
pmu2bidsphysio --infile <pmufiles> --bidsprefix <Prefix>
edf2bidsphysio --infile <edffile> --bidsprefix <Prefix>

How to use in your own Python program

After installing the module using pip (see above ), you can use it in your own Python program this way:

from bidsphysio.dcm2bids import dcm2bidsphysio
dcm2bidsphysio.dcm2bids( dicom_file, prefix )
# or:
dcm2bidsphysio.dcm2bids( [log_files], prefix )

or:

from bidsphysio.acq2bids import acq2bidsphysio
acq2bidsphysio.acq2bids( [acq_files], prefix )

or:

from bidsphysio.pmu2bids import pmu2bidsphysio
pmu2bidsphysio.pmu2bids( [pmu_files], prefix )

or:

from bidsphysio.edf2bids import edf2bidsphysio
edf2bidsphysio.edf2bids( edf_file, prefix )
edf2bidsphysio.edfevents2bids( edf_file, prefix )