data2bids.m

function [cfg] = data2bids(cfg, varargin)

% DATA2BIDS is a helper function to convert MRI, MEG, EEG, iEEG or NIRS data to the
% Brain Imaging Data Structure. The overall idea is that you write a MATLAB script in
% which you call this function multiple times, once for each individually recorded
% data file (or data set). It will write the corresponding sidecar JSON and TSV files
% for each data file.
%
% Use as
%   data2bids(cfg)
% or as
%   data2bids(cfg, data)
%
% The first input argument 'cfg' is the configuration structure, which contains the
% details for the (meta)data and which specifies the sidecar files you want to write.
% The optional 'data' argument corresponds to preprocessed raw data according to
% FT_DATAYPE_RAW or an anatomical MRI according to FT_DATAYPE_VOLUME. The optional
% data input argument allows you to write preprocessed electrophysiological data
% and/or realigned and defaced anatomical MRI to disk.
%
% The configuration structure should contains
%   cfg.method       = string, can be 'decorate', 'copy' or 'convert', see below (default is automatic)
%   cfg.dataset      = string, filename of the input data
%   cfg.outputfile   = string, optional filename for the output data (default is automatic)
%   cfg.writejson    = string, 'yes', 'replace', 'merge' or 'no' (default = 'yes')
%   cfg.writetsv     = string, 'yes', 'replace', 'merge' or 'no' (default = 'yes')
%
% This function starts from existing data file on disk or from a FieldTrip compatible
% data structure in MATLAB memory that is passed as the second input argument.
% Depending on cfg.method it will add the sidecar files, copy the dataset and add
% sidecar files, or convert the dataset and add the sidecar files. Each of the
% methods is discussed here.
%
% DECORATE - data2bids will read the header details and events from the data and write
% the appropriate sidecar files alongside the existing dataset. You would use this to
% obtain the sidecar files for data files that are already in the BIDS organization.
%
% CONVERT - data2bids will read the input data (or use the specified input data) and
% write it to a new output file that is BIDS compliant. The output format is NIfTI
% for MRI data, and BrainVision for EEG and iEEG. Note that MEG data files are stored
% in BIDS in their native format and this function will NOT convert them for you.
%
% COPY - data2bids will copy the data from the input data file to the output data
% file, which renames it, but does not change its content. Furthermore, it will read
% the header details and events from the data and construct the appropriate sidecar
% files.
%
% Although you can explicitly specify cfg.outputfile yourself, it is recommended to
% use the following configuration options. This results in a BIDS compliant output
% directory and file name. With these options data2bids will also write, or if
% already present update the participants.tsv and scans.tsv files.
%   cfg.bidsroot                = string, top level directory for the BIDS output
%   cfg.sub                     = string, subject name
%   cfg.ses                     = string, optional session name
%   cfg.run                     = number, optional
%   cfg.task                    = string, task name is required for functional data
%   cfg.datatype                = string, can be any of 'FLAIR', 'FLASH', 'PD', 'PDT2', 'PDmap', 'T1map', 'T1rho', 'T1w', 'T2map', 'T2star', 'T2w', 'angio', 'audio', 'bold', 'bval', 'bvec', 'channels', 'coordsystem', 'defacemask', 'dwi', 'eeg', 'emg', 'epi', 'events', 'exg', 'eyetracker', 'fieldmap', 'headshape', 'ieeg', 'inplaneT1', 'inplaneT2', 'magnitude', 'magnitude1', 'magnitude2', 'meg', 'motion', 'nirs', 'phase1', 'phase2', 'phasediff', 'photo', 'physio', 'sbref', 'stim', 'video'
%   cfg.acq                     = string
%   cfg.ce                      = string
%   cfg.rec                     = string
%   cfg.dir                     = string
%   cfg.mod                     = string
%   cfg.echo                    = string
%   cfg.proc                    = string
%   cfg.tracksys                = string
%   cfg.space                   = string
%   cfg.desc                    = string
%
% When specifying the output directory in cfg.bidsroot, you can also specify
% additional information to be added as extra columns in the participants.tsv and
% scans.tsv files. For example:
%   cfg.participants.age        = scalar
%   cfg.participants.sex        = string, 'm' or 'f'
%   cfg.scans.acq_time          = string, should be formatted according to RFC3339 as '2019-05-22T15:13:38'
%   cfg.sessions.acq_time       = string, should be formatted according to RFC3339 as '2019-05-22T15:13:38'
%   cfg.sessions.pathology      = string, recommended when different from healthy
% In case any of these values is specified as empty (i.e. []) or as nan, it will be
% written to the tsv file as 'n/a'.
%
% For anatomical and functional MRI data you can specify cfg.dicomfile to read the
% detailed MRI scanner and sequence details from the header of that DICOM file. This
% will be used to fill in the details of the corresponding JSON file.
%   cfg.dicomfile               = string, filename of a matching DICOM file for header details (default = [])
%   cfg.deface                  = string, 'yes' or 'no' (default = 'no')
%
% You can specify cfg.events as a Nx3 matrix with the trial definition (see
% FT_DEFINETRIAL) or as a MATLAB table. When specified as table, the first three
% columns containing integer values corresponding to the begsample, endsample and
% offset, the additional colums can be of another type and have any name. If you do
% not specify the trial definition, the events will be read from the MEG/EEG/iEEG
% dataset. Events from the trial definition or from the data will be written to
% events.tsv.
%   cfg.events                  = trial definition (see FT_DEFINETRIAL) or event structure (see FT_READ_EVENT)
%
% If NBS Presentation was used in combination with another functional data type, you
% can specify cfg.presentationfile with the name of the presentation log file, which
% will be aligned with the data based on triggers (MEG/EEG/iEEG) or based on the
% volumes (fMRI). Events from the presentation log file will also be written to
% events.tsv. To indicate how triggers (in MEG/EEG/iEEG) or volumes (in fMRI) match
% the presentation events, you should specify the mapping between them.
%   cfg.presentationfile        = string, optional filename for the presentation log file
%   cfg.trigger.eventtype       = string (default = [])
%   cfg.trigger.eventvalue      = string or number
%   cfg.trigger.skip            = 'last'/'first'/'none'
%   cfg.presentation.eventtype  = string (default = [])
%   cfg.presentation.eventvalue = string or number
%   cfg.presentation.skip       = 'last'/'first'/'none'
%
% For EEG and iEEG data you can specify an electrode definition according to
% FT_DATATYPE_SENS as an "elec" field in the input data, or you can specify it as
% cfg.elec or you can specify a filename with electrode information.
%   cfg.elec                    = structure with electrode positions or filename, see FT_READ_SENS
%
% For NIRS data you can specify an optode definition according to
% FT_DATATYPE_SENS as an "opto" field in the input data, or you can specify
% it as cfg.opto or you can specify a filename with optode information.
%   cfg.opto                    = structure with optode positions or filename,see FT_READ_SENS
%
% General BIDS options that apply to all data types are
%   cfg.InstitutionName             = string
%   cfg.InstitutionAddress          = string
%   cfg.InstitutionalDepartmentName = string
%   cfg.Manufacturer                = string
%   cfg.ManufacturersModelName      = string
%   cfg.DeviceSerialNumber          = string
%   cfg.SoftwareVersions            = string
%
% If you specify cfg.bidsroot, this function will also write the dataset_description.json
% file. You can specify the following fields
%   cfg.dataset_description.writesidecar        = string
%   cfg.dataset_description.Name                = string
%   cfg.dataset_description.BIDSVersion         = string
%   cfg.dataset_description.License             = string
%   cfg.dataset_description.Authors             = string or cell-array of strings
%   cfg.dataset_description.Acknowledgements    = string
%   cfg.dataset_description.HowToAcknowledge    = string
%   cfg.dataset_description.Funding             = string or cell-array of strings
%   cfg.dataset_description.ReferencesAndLinks  = string or cell-array of strings
%   cfg.dataset_description.DatasetDOI          = string
%
% General BIDS options that apply to all functional data types are
%   cfg.TaskName                    = string
%   cfg.TaskDescription             = string
%   cfg.Instructions                = string
%   cfg.CogAtlasID                  = string
%   cfg.CogPOID                     = string
%
% There are more BIDS options for the mri/meg/eeg/ieeg data type specific sidecars.
% Rather than listing them all here, please open this function in the MATLAB editor,
% and scroll down a bit to see what those are. In general the information in the JSON
% files is specified by a field that is specified in CamelCase
%   cfg.mri.SomeOption              = string, please check the MATLAB code
%   cfg.meg.SomeOption              = string, please check the MATLAB code
%   cfg.eeg.SomeOption              = string, please check the MATLAB code
%   cfg.ieeg.SomeOption             = string, please check the MATLAB code
%   cfg.nirs.SomeOption             = string, please check the MATLAB code
%   cfg.coordsystem.someoption      = string, please check the MATLAB code
% The information for TSV files is specified with a column header in lowercase or
% snake_case and represents a list of items
%   cfg.channels.someoption         = cell-array, please check the MATLAB code
%   cfg.events.someoption           = cell-array, please check the MATLAB code
%   cfg.electrodes.someoption       = cell-array, please check the MATLAB code
%   cfg.optodes.someoption          = cell-array, please check the MATLAB code
%
% The implementation in this function aims to correspond to the latest BIDS version.
% See https://bids-specification.readthedocs.io/ for the full specification
% and http://bids.neuroimaging.io/ for further details.
%
% See also FT_DATAYPE_RAW, FT_DATAYPE_VOLUME, FT_DATATYPE_SENS, FT_DEFINETRIAL,
% FT_PREPROCESSING, FT_READ_MRI, FT_READ_EVENT

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Undocumented options exist for converting some other data types to BIDS:
% - emg
% - exg
% - audio
% - video
% - eyetracking
% - motion
% - physio
% - stim
%
% Most of these data types are currently (Oct 2020) not supported in the BIDS
% specification, but this function converts them in a very similar way as the
% officially supported data types.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% Copyright (C) 2018-2020, Robert Oostenveld
%
% This file is part of FieldTrip, see http://www.fieldtriptoolbox.org
% for the documentation and details.
%
%    FieldTrip is free software: you can redistribute it and/or modify
%    it under the terms of the GNU General Public License as published by
%    the Free Software Foundation, either version 3 of the License, or
%    (at your option) any later version.
%
%    FieldTrip is distributed in the hope that it will be useful,
%    but WITHOUT ANY WARRANTY; without even the implied warranty of
%    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
%    GNU General Public License for more details.
%
%    You should have received a copy of the GNU General Public License
%    along with FieldTrip. If not, see <http://www.gnu.org/licenses/>.
%
% $Id$

% these are used by the ft_preamble/ft_postamble function and scripts
ft_revision = '$Id$';
ft_nargin   = nargin;
ft_nargout  = nargout;

% do the general setup of the function
ft_defaults
ft_preamble init
ft_preamble debug

if ft_abort
    % do not continue function execution in case the outputfile is present and the user indicated to keep it
    return
end

% ensure backward compatibility
cfg = ft_checkconfig(cfg, 'renamed', {'participant', 'participants'}); % this was wrong in the documentation
cfg = ft_checkconfig(cfg, 'renamed', {'anat', 'mri'});
cfg = ft_checkconfig(cfg, 'renamedval', {'native', 'no', 'convert'});
cfg = ft_checkconfig(cfg, 'renamedval', {'native', 'yes', 'copy'});
cfg = ft_checkconfig(cfg, 'renamed', {'native', 'method'});
cfg = ft_checkconfig(cfg, 'renamed', {'mri.deface', 'deface'});
cfg = ft_checkconfig(cfg, 'renamed', {'mri.writesidecar', 'writejson'});
cfg = ft_checkconfig(cfg, 'renamed', {'meg.writesidecar', 'writejson'});
cfg = ft_checkconfig(cfg, 'renamed', {'eeg.writesidecar', 'writejson'});
cfg = ft_checkconfig(cfg, 'renamed', {'ieeg.writesidecar', 'writejson'});
cfg = ft_checkconfig(cfg, 'renamed', {'events.writesidecar', 'writejson'});
cfg = ft_checkconfig(cfg, 'renamed', {'channels.writesidecar', 'writejson'});
cfg = ft_checkconfig(cfg, 'renamed', {'electrodes.writesidecar', 'writejson'});
cfg = ft_checkconfig(cfg, 'renamed', {'coordsystem.writesidecar', 'writejson'});
cfg = ft_checkconfig(cfg, 'renamed', {'event', 'events'}); % cfg.event is used elsewhere in FieldTrip, but here it should be cfg.events with an s

% prevent some common errors
cfg = ft_checkconfig(cfg, 'forbidden', {'acq_time'});                 % this should be in cfg.scans or in cfg.sessions
cfg = ft_checkconfig(cfg, 'forbidden', {'scan', 'session', 'event'}); % these should end with an "s"

% get the options and set the defaults
cfg.method                  = ft_getopt(cfg, 'method');                     % default is handled below
cfg.dataset                 = ft_getopt(cfg, 'dataset');
cfg.feedback                = ft_getopt(cfg, 'feedback', 'yes');
cfg.outputfile              = ft_getopt(cfg, 'outputfile');                 % default is handled below
cfg.presentationfile        = ft_getopt(cfg, 'presentationfile');           % full path to the NBS presentation log file, it will be read and parsed using FT_READ_EVENT
cfg.presentation            = ft_getopt(cfg, 'presentation');
cfg.presentation.eventtype  = ft_getopt(cfg.presentation, 'eventtype');
cfg.presentation.eventvalue = ft_getopt(cfg.presentation, 'eventvalue');
cfg.presentation.skip       = ft_getopt(cfg.presentation, 'skip',  'last');  % this is a sensible default for fMRI, for MEG one should probably do 'none'
cfg.trigger                 = ft_getopt(cfg, 'trigger');
cfg.trigger.eventtype       = ft_getopt(cfg.trigger, 'eventtype');
cfg.trigger.eventvalue      = ft_getopt(cfg.trigger, 'eventvalue');
cfg.trigger.skip            = ft_getopt(cfg.trigger, 'skip',  'none');

% these are used to construct the directory and file name
cfg.bidsroot  = ft_getopt(cfg, 'bidsroot');
cfg.sub       = ft_getopt(cfg, 'sub');
cfg.ses       = ft_getopt(cfg, 'ses');
cfg.task      = ft_getopt(cfg, 'task');
cfg.tracksys  = ft_getopt(cfg, 'tracksys');
cfg.acq       = ft_getopt(cfg, 'acq');
cfg.ce        = ft_getopt(cfg, 'ce');
cfg.rec       = ft_getopt(cfg, 'rec');
cfg.dir       = ft_getopt(cfg, 'dir');
cfg.run       = ft_getopt(cfg, 'run');
cfg.mod       = ft_getopt(cfg, 'mod');
cfg.echo      = ft_getopt(cfg, 'echo');
cfg.proc      = ft_getopt(cfg, 'proc');
cfg.space     = ft_getopt(cfg, 'space');
cfg.desc      = ft_getopt(cfg, 'desc');
cfg.datatype  = ft_getopt(cfg, 'datatype');

% do a sanity check on the fields that form the filename as key-value pair
fn = {'sub', 'ses', 'task', 'tracksys', 'acq', 'ce', 'rec', 'dir', 'run', 'mod', 'echo', 'proc', 'space', 'desc'};

for i=1:numel(fn)
    if ischar(cfg.(fn{i})) && any(cfg.(fn{i})=='-')
        ft_error('the field cfg.%s cannot contain a "-"', fn{i});
    end
end

if isempty(cfg.datatype)
    modality = {'meg', 'eeg', 'ieeg', 'emg', 'exg', 'audio', 'video', 'eyetracker', 'physio', 'stim', 'motion', 'nirs'};
    for i=1:numel(modality)
        if isfield(cfg, modality{i}) && ~isempty(cfg.(modality{i}))
            % the user specified modality-specific options, assume that the datatype matches
            cfg.datatype = modality{i};
            ft_notice('assuming that the datatype is %s', cfg.datatype);
            continue
        end
    end % for each modality
end

cfg.dicomfile     = ft_getopt(cfg, 'dicomfile');                      % get header details from the specified DICOM files
cfg.deface        = ft_getopt(cfg, 'deface', 'no');                   % whether to deface the anatomical MRI
cfg.writejson     = ft_getopt(cfg, 'writejson', 'merge');             % whether to write the json file
cfg.writetsv      = ft_getopt(cfg, 'writetsv', 'merge');              % whether to write the tsv file

cfg.mri           = ft_getopt(cfg, 'mri');
cfg.meg           = ft_getopt(cfg, 'meg');
cfg.eeg           = ft_getopt(cfg, 'eeg');
cfg.ieeg          = ft_getopt(cfg, 'ieeg');
cfg.emg           = ft_getopt(cfg, 'emg');
cfg.exg           = ft_getopt(cfg, 'exg');
cfg.nirs          = ft_getopt(cfg, 'nirs');
cfg.audio         = ft_getopt(cfg, 'audio');
cfg.video         = ft_getopt(cfg, 'video');
cfg.eyetracker    = ft_getopt(cfg, 'eyetracker');
cfg.physio        = ft_getopt(cfg, 'physio');
cfg.stim          = ft_getopt(cfg, 'stim');
cfg.motion        = ft_getopt(cfg, 'motion');
cfg.channels      = ft_getopt(cfg, 'channels');
cfg.electrodes    = ft_getopt(cfg, 'electrodes');
cfg.optodes       = ft_getopt(cfg, 'optodes');
cfg.events        = ft_getopt(cfg, 'events');     % this can contain the trial definition as Nx3 array, as table, or an event structure
cfg.coordsystem   = ft_getopt(cfg, 'coordsystem');

%% Dataset description

cfg.dataset_description                     = ft_getopt(cfg, 'dataset_description'                       );
cfg.dataset_description.writesidecar        = ft_getopt(cfg.dataset_description, 'writesidecar', 'yes'   );
cfg.dataset_description.Name                = ft_getopt(cfg.dataset_description, 'Name'                  ); % REQUIRED. Name of the dataset.
cfg.dataset_description.BIDSVersion         = ft_getopt(cfg.dataset_description, 'BIDSVersion', '1.8'    ); % REQUIRED. The version of the BIDS standard that was used.
cfg.dataset_description.DatasetType         = ft_getopt(cfg.dataset_description, 'DatasetType', 'raw'    ); % RECOMMENDED. The interpretaton of the dataset. MUST be one of "raw" or "derivative". For backwards compatibility, the default value is "raw".
cfg.dataset_description.License             = ft_getopt(cfg.dataset_description, 'License'               ); % RECOMMENDED. What license is this dataset distributed under? The use of license name abbreviations is suggested for specifying a license. A list of common licenses with suggested abbreviations can be found in Appendix II.
cfg.dataset_description.Authors             = ft_getopt(cfg.dataset_description, 'Authors'               ); % OPTIONAL. List of individuals who contributed to the creation/curation of the dataset.
cfg.dataset_description.Acknowledgements    = ft_getopt(cfg.dataset_description, 'Acknowledgements'      ); % OPTIONAL. Text acknowledging contributions of individuals or institutions beyond those listed in Authors or Funding.
cfg.dataset_description.HowToAcknowledge    = ft_getopt(cfg.dataset_description, 'HowToAcknowledge'      ); % OPTIONAL. Instructions how researchers using this dataset should acknowledge the original authors. This field can also be used to define a publication that should be cited in publications that use the dataset.
cfg.dataset_description.Funding             = ft_getopt(cfg.dataset_description, 'Funding'               ); % OPTIONAL. List of sources of funding (grant numbers)
cfg.dataset_description.EthicsApprovals     = ft_getopt(cfg.dataset_description, 'EthicsApprovals'       ); % OPTIONAL. List of ethics committee approvals of the research protocols and/or protocol identifiers.
cfg.dataset_description.ReferencesAndLinks  = ft_getopt(cfg.dataset_description, 'ReferencesAndLinks'    ); % OPTIONAL. List of references to publication that contain information on the dataset, or links.
cfg.dataset_description.DatasetDOI          = ft_getopt(cfg.dataset_description, 'DatasetDOI'            ); % OPTIONAL. The Document Object Identifier of the dataset (not the corresponding paper).

%% Generic fields for all data types
cfg.TaskName                          = ft_getopt(cfg, 'TaskName'                    ); % REQUIRED. Name of the task (for resting state use the "rest" prefix). Different Tasks SHOULD NOT have the same name. The Task label is derived from this field by removing all non alphanumeric ([a-zA-Z0-9]) characters.
cfg.TaskDescription                   = ft_getopt(cfg, 'TaskDescription'             ); % OPTIONAL. Description of the task.
cfg.Instructions                      = ft_getopt(cfg, 'Instructions'                ); % OPTIONAL. Text of the instructions given to participants before the scan. This is not only important for behavioral or cognitive tasks but also in resting state paradigms (e.g. to distinguish between eyes open and eyes closed).
cfg.CogAtlasID                        = ft_getopt(cfg, 'CogAtlasID'                  ); % OPTIONAL. URL of the corresponding "Cognitive Atlas term that describes the task (e.g. Resting State with eyes closed ""http://www.cognitiveatlas.org/term/id/trm_54e69c642d89b""")
cfg.CogPOID                           = ft_getopt(cfg, 'CogPOID'                     ); % OPTIONAL. URL of the corresponding "CogPO term that describes the task (e.g. Rest "http://wiki.cogpo.org/index.php?title=Rest")
cfg.Manufacturer                      = ft_getopt(cfg, 'Manufacturer'                ); % OPTIONAL. Manufacturer of the recording system ("CTF", ""Neuromag/Elekta"", ""4D/BTi"", ""KIT/Yokogawa"", ""ITAB"", "KRISS", "Other")
cfg.ManufacturersModelName            = ft_getopt(cfg, 'ManufacturersModelName'      ); % OPTIONAL. Manufacturer's designation of the model (e.g. "CTF-275"). See "Appendix VII" with preferred names
cfg.DeviceSerialNumber                = ft_getopt(cfg, 'DeviceSerialNumber'          ); % OPTIONAL. The serial number of the equipment that produced the composite instances. A pseudonym can also be used to prevent the equipment from being identifiable, as long as each pseudonym is unique within the dataset.
cfg.SoftwareVersions                  = ft_getopt(cfg, 'SoftwareVersions'            ); % OPTIONAL. Manufacturer's designation of the acquisition software.
cfg.InstitutionName                   = ft_getopt(cfg, 'InstitutionName'             ); % OPTIONAL. The name of the institution in charge of the equipment that produced the composite instances.
cfg.InstitutionAddress                = ft_getopt(cfg, 'InstitutionAddress'          ); % OPTIONAL. The address of the institution in charge of the equipment that produced the composite instances.
cfg.InstitutionalDepartmentName       = ft_getopt(cfg, 'InstitutionalDepartmentName' ); % The department in the institution in charge of the equipment that produced the composite instances. Corresponds to DICOM Tag 0008, 1040 "Institutional Department Name".

%% MR Scanner Hardware
cfg.mri.StationName                   = ft_getopt(cfg.mri, 'StationName'                    ); % Institution defined name of the machine that produced the composite instances. Corresponds to DICOM Tag 0008, 1010 "Station Name"
cfg.mri.HardcopyDeviceSoftwareVersion = ft_getopt(cfg.mri, 'HardcopyDeviceSoftwareVersion'  ); % (Deprecated) Manufacturer's designation of the software of the device that created this Hardcopy Image (the printer). Corresponds to DICOM Tag 0018, 101A "Hardcopy Device Software Version".
cfg.mri.MagneticFieldStrength         = ft_getopt(cfg.mri, 'MagneticFieldStrength'          ); % Nominal field strength of MR magnet in Tesla. Corresponds to DICOM Tag 0018,0087 "Magnetic Field Strength" .
cfg.mri.ReceiveCoilName               = ft_getopt(cfg.mri, 'ReceiveCoilName'                ); % Information describing the receiver coil. Corresponds to DICOM Tag 0018, 1250 "Receive Coil Name", although not all vendors populate that DICOM Tag, in which case this field can be derived from an appropriate private DICOM field.
cfg.mri.ReceiveCoilActiveElements     = ft_getopt(cfg.mri, 'ReceiveCoilActiveElements'      ); % Information describing the active/selected elements of the receiver coil.  This doesn't correspond to a tag in the DICOM ontology. The vendor-defined terminology for active coil elements can go in this field. As an example, for Siemens, coil channels are typically not activated/selected individually, but rather  in pre-defined selectable "groups" of individual channels, and the list of the  "groups" of elements that are active/selected in any given scan populates  the "Coil String" entry in Siemen's private DICOM fields (e.g., "HEA;HEP" for the Siemens standard 32 ch coil when both the anterior and posterior groups are activated). This is a flexible field that can be used as most appropriate for a given vendor and coil to define the "active" coil elements. Since individual scans can sometimes not have the intended coil elements selected, it is preferable for this field to be populated directly from the DICOM for each individual scan, so that it can be used as a mechanism for checking that a given scan was collected with the intended coil elements selected.
cfg.mri.GradientSetType               = ft_getopt(cfg.mri, 'GradientSetType'                ); % It should be possible to infer the gradient coil from the scanner model. If not,e.g. because of a custom upgrade or use of a gradient insert set, then the specifications of the actual gradient coil should be reported independently.
cfg.mri.MRTransmitCoilSequence        = ft_getopt(cfg.mri, 'MRTransmitCoilSequence'         ); % This is a relevant field if a non-standard transmit coil is used. Corresponds to DICOM Tag 0018, 9049 "MR Transmit Coil Sequence".
cfg.mri.MatrixCoilMode                = ft_getopt(cfg.mri, 'MatrixCoilMode'                 ); % (If used) A method for reducing the number of independent channels by combining in analog the signals from multiple coil elements. There are typically different default modes when using un-accelerated or accelerated (e.g. GRAPPA, SENSE) imaging.
cfg.mri.CoilCombinationMethod         = ft_getopt(cfg.mri, 'CoilCombinationMethod'          ); % Almost all fMRI studies using phased-array coils use root-sum-of-squares (rSOS) combination, but other methods exist. The image reconstruction is changed by the coil combination method (as for the matrix coil mode above), so anything non-standard should be reported.

%% MR Sequence Specifics
cfg.mri.PulseSequenceType             = ft_getopt(cfg.mri, 'PulseSequenceType'              ); % A general description of the pulse sequence used for the scan (i.e. MPRAGE, Gradient Echo EPI, Spin Echo EPI, Multiband gradient echo EPI).
cfg.mri.ScanningSequence              = ft_getopt(cfg.mri, 'ScanningSequence'               ); % Description of the type of data acquired. Corresponds to DICOM Tag 0018, 0020 "Sequence Sequence".
cfg.mri.SequenceVariant               = ft_getopt(cfg.mri, 'SequenceVariant'                ); % Variant of the ScanningSequence. Corresponds to DICOM Tag 0018, 0021 "Sequence Variant".
cfg.mri.ScanOptions                   = ft_getopt(cfg.mri, 'ScanOptions'                    ); % Parameters of ScanningSequence. Corresponds to DICOM Tag 0018, 0022 "Scan Options".
cfg.mri.SequenceName                  = ft_getopt(cfg.mri, 'SequenceName'                   ); % Manufacturer's designation of the sequence name. Corresponds to DICOM Tag 0018, 0024 "Sequence Name".
cfg.mri.PulseSequenceDetails          = ft_getopt(cfg.mri, 'PulseSequenceDetails'           ); % Information beyond pulse sequence type that identifies the specific pulse sequence used (i.e. "Standard Siemens Sequence distributed with the VB17 software," "Siemens WIP ### version #.##," or "Sequence written by X using a version compiled on MM/DD/YYYY").
cfg.mri.NonlinearGradientCorrection   = ft_getopt(cfg.mri, 'NonlinearGradientCorrection'    ); % Boolean stating if the image saved  has been corrected for gradient nonlinearities by the scanner sequence.

%% MR In-Plane Spatial Encoding
cfg.mri.NumberShots                   = ft_getopt(cfg.mri, 'NumberShots'                    ); % The number of RF excitations need to reconstruct a slice or volume. Please mind that  this is not the same as Echo Train Length which denotes the number of lines of k-space collected after an excitation.
cfg.mri.ParallelReductionFactorInPlan = ft_getopt(cfg.mri, 'ParallelReductionFactorInPlane' ); % The parallel imaging (e.g, GRAPPA) factor. Use the denominator of the fraction of k-space encoded for each slice. For example, 2 means half of k-space is encoded. Corresponds to DICOM Tag 0018, 9069 "Parallel Reduction Factor In-plane".
cfg.mri.ParallelAcquisitionTechnique  = ft_getopt(cfg.mri, 'ParallelAcquisitionTechnique'   ); % The type of parallel imaging used (e.g. GRAPPA, SENSE). Corresponds to DICOM Tag 0018, 9078 "Parallel Acquisition Technique".
cfg.mri.PartialFourier                = ft_getopt(cfg.mri, 'PartialFourier'                 ); % The fraction of partial Fourier information collected. Corresponds to DICOM Tag 0018, 9081 "Partial Fourier".
cfg.mri.PartialFourierDirection       = ft_getopt(cfg.mri, 'PartialFourierDirection'        ); % The direction where only partial Fourier information was collected. Corresponds to DICOM Tag 0018, 9036 "Partial Fourier Direction".
cfg.mri.PhaseEncodingDirection        = ft_getopt(cfg.mri, 'PhaseEncodingDirection'         ); % Possible values = [];                     % "i", "j", "k", "i-", "j-", "k-". The letters "i", "j", "k" correspond to the first, second and third axis of the data in the NIfTI file. The polarity of the phase encoding is assumed to go from zero index to maximum index unless "-" sign is present (then the order is reversed - starting from the highest index instead of zero). PhaseEncodingDirection is defined as the direction along which phase is was modulated which may result in visible distortions. Note that this is not the same as the DICOM term InPlanePhaseEncodingDirection which can have "ROW" or "COL" values. This parameter is REQUIRED if corresponding fieldmap data is present or when using multiple runs with different phase encoding directions (which can be later used for field inhomogeneity correction).
cfg.mri.EffectiveEchoSpacing          = ft_getopt(cfg.mri, 'EffectiveEchoSpacing'           ); % The "effective" sampling interval, specified in seconds, between lines in the phase-encoding direction, defined based on the size of the reconstructed image in the phase direction.  It is frequently, but incorrectly, referred to as  "dwell time" (see DwellTime parameter below for actual dwell time).  It is  required for unwarping distortions using field maps. Note that beyond just in-plane acceleration, a variety of other manipulations to the phase encoding need to be accounted for properly, including partial fourier, phase oversampling, phase resolution, phase field-of-view and interpolation. This parameter is REQUIRED if corresponding fieldmap data is present.
cfg.mri.TotalReadoutTime              = ft_getopt(cfg.mri, 'TotalReadoutTime'               ); % This is actually the "effective" total readout time , defined as the readout duration, specified in seconds, that would have generated data with the given level of distortion.  It is NOT the actual, physical duration of the readout train.  If EffectiveEchoSpacing has been properly computed, it is just EffectiveEchoSpacing * (ReconMatrixPE - 1). . This parameter is REQUIRED if corresponding "field/distortion" maps acquired with opposing phase encoding directions are present  (see 8.9.4).

%% MR Timing Parameters
cfg.mri.EchoTime                      = ft_getopt(cfg.mri, 'EchoTime'                       ); % The echo time (TE) for the acquisition, specified in seconds. This parameter is REQUIRED if corresponding fieldmap data is present or the data comes from a multi echo sequence. Corresponds to DICOM Tag 0018, 0081 "Echo Time"  (please note that the DICOM term is in milliseconds not seconds).
cfg.mri.InversionTime                 = ft_getopt(cfg.mri, 'InversionTime'                  ); % The inversion time (TI) for the acquisition, specified in seconds. Inversion time is the time after the middle of inverting RF pulse to middle of excitation pulse to detect the amount of longitudinal magnetization. Corresponds to DICOM Tag 0018, 0082 "Inversion Time"  (please note that the DICOM term is in milliseconds not seconds).
cfg.mri.SliceTiming                   = ft_getopt(cfg.mri, 'SliceTiming'                    ); % The time at which each slice was acquired within each volume (frame) of  the acquisition.  Slice timing is not slice order -- rather, it  is a list of times (in JSON format) containing the time (in seconds) of each slice acquisition in relation to the beginning of volume acquisition.  The list goes through the slices along the slice axis in the slice encoding dimension (see below). Note that to ensure the proper interpretation of the SliceTiming field, it is important to check if the (optional) SliceEncodingDirection exists. In particular,  if SliceEncodingDirection is negative, the entries in SliceTiming are defined in reverse order with respect to the slice axis (i.e., the final entry in the SliceTiming list is the time of acquisition of slice 0). This parameter is REQUIRED for sparse sequences that do not have the DelayTime field set. In addition without this parameter slice time correction will not be possible.
cfg.mri.SliceEncodingDirection        = ft_getopt(cfg.mri, 'SliceEncodingDirection'         ); % Possible values "i", "j", "k", "i-", "j-", "k-" (the axis of the NIfTI data along which slices were acquired, and the direction in which SliceTiming is  defined with respect to). "i", "j", "k" identifiers correspond to the first, second and third axis of the data in the NIfTI file. A "-" sign indicates that the contents of SliceTiming are defined in reverse order -- that is, the first entry corresponds to the slice with the largest index, and the final entry corresponds to slice index zero. When present ,the axis defined by SliceEncodingDirection  needs to be consistent with the "slice_dim" field in the NIfTI header. When absent, the entries in SliceTiming must be in the order of increasing slice index as defined by the NIfTI header.
cfg.mri.DwellTime                     = ft_getopt(cfg.mri, 'DwellTime'                      ); % Actual dwell time (in seconds) of the receiver per point in the readout direction, including any oversampling.  For Siemens, this corresponds to DICOM field (0019,1018) (in ns).   This value is necessary for the (optional) readout distortion correction of anatomicals in the HCP Pipelines.  It also usefully provides a handle on the readout bandwidth, which isn't captured in the other metadata tags.  Not to be confused with "EffectiveEchoSpacing", and the frequent mislabeling of echo spacing (which is spacing in the phase encoding direction) as "dwell time" (which is spacing in the readout direction).

%% MR RF & Contrast
cfg.mri.FlipAngle                     = ft_getopt(cfg.mri, 'FlipAngle'                      ); % Flip angle for the acquisition, specified in degrees. Corresponds to = [];                     % DICOM Tag 0018, 1314 "Flip Angle".
cfg.mri.MultibandAccelerationFactor   = ft_getopt(cfg.mri, 'MultibandAccelerationFactor'    ); % RECOMMENDED. The multiband factor, for multiband acquisitions.
cfg.mri.NegativeContrast              = ft_getopt(cfg.mri, 'NegativeContrast'               ); % OPTIONAL. Boolean (true or false) value specifying whether increasing voxel intensity (within sample voxels) denotes a decreased value with respect to the contrast suffix. This is commonly the case when Cerebral Blood Volume is estimated via usage of a contrast agent in conjunction with a T2* weighted acquisition protocol.

%% MR Slice Acceleration
cfg.mri.MultibandAccelerationFactor   = ft_getopt(cfg.mri, 'MultibandAccelerationFactor'    ); % The multiband factor, for multiband acquisitions.

%% Anatomical landmarks, useful for multimodaltimodal co-registration with MEG, (S)HeadCoil, TMS,etc
cfg.mri.AnatomicalLandmarkCoordinates = ft_getopt(cfg.mri, 'AnatomicalLandmarkCoordinates'  ); % Key:value pairs of any number of additional anatomical landmarks and their coordinates in voxel units (where first voxel has index 0,0,0) relative to the associated anatomical MRI, (e.g. {"AC" = []; % [127,119,149], "PC" = []; % [128,93,141], "IH" = []; % [131,114,206]}, or {"NAS" = []; % [127,213,139], "LPA" = []; % [52,113,96], "RPA" = []; % [202,113,91]}).

%% MR Anatomical scan information
cfg.mri.ContrastBolusIngredient       = ft_getopt(cfg.mri, 'ContrastBolusIngredient'        ); % OPTIONAL. Active ingredient of agent. Values MUST be one of: IODINE, GADOLINIUM, CARBON DIOXIDE, BARIUM, XENON Corresponds to DICOM Tag 0018,1048.

%% MR Functional scan information
cfg.mri.RepetitionTime                = ft_getopt(cfg.mri, 'RepetitionTime'                 ); % REQUIRED. The time in seconds between the beginning of an acquisition of one volume and the beginning of acquisition of the volume following it (TR). Please note that this definition includes time between scans (when no data has been acquired) in case of sparse acquisition schemes. This value needs to be consistent with the pixdim[4] field (after accounting for units stored in xyzt_units field) in the NIfTI header. This field is mutually exclusive with VolumeTiming and is derived from DICOM Tag 0018, 0080 and converted to seconds.
cfg.mri.VolumeTiming                  = ft_getopt(cfg.mri, 'VolumeTiming'                   ); % REQUIRED. The time at which each volume was acquired during the acquisition. It is described using a list of times (in JSON format) referring to the onset of each volume in the BOLD series. The list must have the same length as the BOLD series, and the values must be non-negative and monotonically increasing. This field is mutually exclusive with RepetitionTime and DelayTime. If defined, this requires acquisition time (TA) be defined via either SliceTiming or AcquisitionDuration be defined.

%% MEG specific fields
% Manufacturer and ManufacturersModelName are general
cfg.meg.SamplingFrequency             = ft_getopt(cfg.meg, 'SamplingFrequency'           ); % REQUIRED. Sampling frequency (in Hz) of all the data in the recording, regardless of their type (e.g., 2400)
cfg.meg.PowerLineFrequency            = ft_getopt(cfg.meg, 'PowerLineFrequency'          ); % REQUIRED. Frequency (in Hz) of the power grid at the geographical location of the MEG instrument (i.e. 50 or 60)
cfg.meg.DewarPosition                 = ft_getopt(cfg.meg, 'DewarPosition'               ); % REQUIRED. Position of the dewar during the MEG scan: "upright", "supine" or "degrees" of angle from vertical: for example on CTF systems, upright=15??, supine = 90??.
cfg.meg.SoftwareFilters               = ft_getopt(cfg.meg, 'SoftwareFilters'             ); % REQUIRED. List of temporal and/or spatial software filters applied, orideally key:valuepairsofpre-appliedsoftwarefiltersandtheir parameter values: e.g., {"SSS": {"frame": "head", "badlimit": 7}}, {"SpatialCompensation": {"GradientOrder": Order of the gradient compensation}}. Write "n/a" if no software filters applied.
cfg.meg.DigitizedLandmarks            = ft_getopt(cfg.meg, 'DigitizedLandmarks'          ); % REQUIRED. Boolean ("true" or "false") value indicating whether anatomical landmark points (i.e. fiducials) are contained within this recording.
cfg.meg.DigitizedHeadPoints           = ft_getopt(cfg.meg, 'DigitizedHeadPoints'         ); % REQUIRED. Boolean ("true" or "false") value indicating whether head points outlining the scalp/face surface are contained within this recording.
cfg.meg.MEGChannelCount               = ft_getopt(cfg.meg, 'MEGChannelCount'             ); % OPTIONAL. Number of MEG channels (e.g. 275)
cfg.meg.MEGREFChannelCount            = ft_getopt(cfg.meg, 'MEGREFChannelCount'          ); % OPTIONAL. Number of MEG reference channels (e.g. 23). For systems without such channels (e.g. Neuromag Vectorview), MEGREFChannelCount"=0
cfg.meg.EEGChannelCount               = ft_getopt(cfg.meg, 'EEGChannelCount'             ); % OPTIONAL. Number of EEG channels recorded simultaneously (e.g. 21)
cfg.meg.ECOGChannelCount              = ft_getopt(cfg.meg, 'ECOGChannelCount'            ); % OPTIONAL. Number of ECoG channels
cfg.meg.SEEGChannelCount              = ft_getopt(cfg.meg, 'SEEGChannelCount'            ); % OPTIONAL. Number of SEEG channels
cfg.meg.EOGChannelCount               = ft_getopt(cfg.meg, 'EOGChannelCount'             ); % OPTIONAL. Number of EOG channels
cfg.meg.ECGChannelCount               = ft_getopt(cfg.meg, 'ECGChannelCount'             ); % OPTIONAL. Number of ECG channels
cfg.meg.EMGChannelCount               = ft_getopt(cfg.meg, 'EMGChannelCount'             ); % OPTIONAL. Number of EMG channels
cfg.meg.MiscChannelCount              = ft_getopt(cfg.meg, 'MiscChannelCount'            ); % OPTIONAL. Number of miscellaneous analog channels for auxiliary signals
cfg.meg.TriggerChannelCount           = ft_getopt(cfg.meg, 'TriggerChannelCount'         ); % OPTIONAL. Number of channels for digital (TTL bit level) triggers
cfg.meg.RecordingDuration             = ft_getopt(cfg.meg, 'RecordingDuration'           ); % OPTIONAL. Length of the recording in seconds (e.g. 3600)
cfg.meg.RecordingType                 = ft_getopt(cfg.meg, 'RecordingType'               ); % OPTIONAL. Defines whether the recording is "continuous" or "epoched"; this latter limited to time windows about events of interest (e.g., stimulus presentations, subject responses etc.)
cfg.meg.EpochLength                   = ft_getopt(cfg.meg, 'EpochLength'                 ); % OPTIONAL. Duration of individual epochs in seconds (e.g. 1) in case of epoched data
cfg.meg.ContinuousHeadLocalization    = ft_getopt(cfg.meg, 'ContinuousHeadLocalization'  ); % OPTIONAL. Boolean ("true" or "false") value indicating whether continuous head localisation was performed.
cfg.meg.HeadCoilFrequency             = ft_getopt(cfg.meg, 'HeadCoilFrequency'           ); % OPTIONAL. List of frequencies (in Hz) used by the head localisation coils ("HLC" in CTF systems, "HPI" in Neuromag/Elekta, "COH" in 4D/BTi) that track the subject's head position in the MEG helmet (e.g. [293, 307, 314, 321])
cfg.meg.MaxMovement                   = ft_getopt(cfg.meg, 'MaxMovement'                 ); % OPTIONAL. Maximum head movement (in mm) detected during the recording, as measured by the head localisation coils (e.g., 4.8)
cfg.meg.SubjectArtefactDescription    = ft_getopt(cfg.meg, 'SubjectArtefactDescription'  ); % OPTIONAL. Freeform description of the observed subject artefact and its possible cause (e.g. "Vagus Nerve Stimulator", "non-removable implant"). If this field is set to "n/a", it will be interpreted as absence of major source of artifacts except cardiac and blinks.
cfg.meg.AssociatedEmptyRoom           = ft_getopt(cfg.meg, 'AssociatedEmptyRoom'         ); % OPTIONAL. Relative path in BIDS folder structure to empty-room file associated with the subject's MEG recording. The path needs to use forward slashes instead of backward slashes (e.g. "sub-emptyroom/ses-<label>/meg/sub-emptyroom_ses-<label>_ta sk-noise_run-<label>_meg.ds").
cfg.meg.HardwareFilters               = ft_getopt(cfg.meg, 'HardwareFilters'             ); % RECOMMENDED. List of temporal hardware filters applied. Ideally key:value pairs of pre-applied hardware filters and their parameter values: e.g., {"HardwareFilters": {"Highpass RC filter": {"Half amplitude cutoff (Hz)": 0.0159, "Roll-off": "6dB/Octave"}}}. Write n/a if no hardware filters applied.

%% Specific EEG fields - if recorded with the MEG system
cfg.meg.EEGPlacementScheme            = ft_getopt(cfg.meg, 'EEGPlacementScheme'          ); % OPTIONAL. Placement scheme of EEG electrodes. Either the name of a standardised placement system (e.g., "10-20") or a list of standardised electrode names (e.g. ["Cz", "Pz"]).
cfg.meg.CapManufacturer               = ft_getopt(cfg.meg, 'CapManufacturer'             ); % OPTIONAL. Manufacturer of the EEG cap (e.g. EasyCap)
cfg.meg.CapManufacturersModelName     = ft_getopt(cfg.meg, 'CapManufacturersModelName'   ); % OPTIONAL. Manufacturer’s designation of the EEG cap model (e.g., M10)
cfg.meg.EEGReference                  = ft_getopt(cfg.meg, 'EEGReference'                ); % OPTIONAL. Description of the type of EEG reference used (e.g., M1 for left mastoid, average, or longitudinal bipolar).

%% EEG specific fields
cfg.eeg.EEGReference                  = ft_getopt(cfg.eeg, 'EEGReference'                ); % Description of the type of reference used (common", "average", "DRL", "bipolar" ).  Any specific electrode used as reference should be indicated as such in the channels.tsv file
cfg.eeg.SamplingFrequency             = ft_getopt(cfg.eeg, 'SamplingFrequency'           ); % Sampling frequency (in Hz) of the EEG recording (e.g. 2400)
cfg.eeg.PowerLineFrequency            = ft_getopt(cfg.eeg, 'PowerLineFrequency'          ); % Frequency (in Hz) of the power grid where the EEG is installed (i.e. 50 or 60).
cfg.eeg.SoftwareFilters               = ft_getopt(cfg.eeg, 'SoftwareFilters'             ); % List of temporal software filters applied or ideally  key:value pairs of pre-applied filters and their parameter values
cfg.eeg.CapManufacturer               = ft_getopt(cfg.eeg, 'CapManufacturer'             ); % name of the cap manufacturer
cfg.eeg.CapManufacturersModelName     = ft_getopt(cfg.eeg, 'CapManufacturersModelName'   ); % Manufacturer's designation of the EEG cap model (e.g. "CAPML128", "actiCAP 64Ch Standard-2")
% Manufacturer and ManufacturersModelName are general
cfg.eeg.EEGChannelCount               = ft_getopt(cfg.eeg, 'EEGChannelCount'             ); % Number of EEG channels included in the recording (e.g. 128).
cfg.eeg.ECGChannelCount               = ft_getopt(cfg.eeg, 'ECGChannelCount'             ); % Number of ECG channels included in the recording (e.g. 1).
cfg.eeg.EMGChannelCount               = ft_getopt(cfg.eeg, 'EMGChannelCount'             ); % Number of EMG channels included in the recording (e.g. 2).
cfg.eeg.EOGChannelCount               = ft_getopt(cfg.eeg, 'EOGChannelCount'             ); % Number of EOG channels included in the recording (e.g. 2).
cfg.eeg.MiscChannelCount              = ft_getopt(cfg.eeg, 'MiscChannelCount'            ); % Number of miscellaneous analog channels for auxiliary  signals
cfg.eeg.TriggerChannelCount           = ft_getopt(cfg.eeg, 'TriggerChannelCount'         ); % Number of channels for digital and analog triggers.
cfg.eeg.RecordingDuration             = ft_getopt(cfg.eeg, 'RecordingDuration'           ); % Length of the recording in seconds (e.g. 3600)
cfg.eeg.RecordingType                 = ft_getopt(cfg.eeg, 'RecordingType'               ); % "continuous", "epoched"
cfg.eeg.EpochLength                   = ft_getopt(cfg.eeg, 'EpochLength'                 ); % Duration of individual epochs in seconds (e.g. 1). If recording was continuous, set value to Inf or leave out the field.
cfg.eeg.HeadCircumference             = ft_getopt(cfg.eeg, 'HeadCircumference'           ); % RECOMMENDED. Circumference of the participants head, expressed in cm (e.g., 58).
cfg.eeg.EEGPlacementScheme            = ft_getopt(cfg.eeg, 'EEGPlacementScheme'          ); % Placement scheme of the EEG electrodes. Either the name of a placement system (e.g. "10-20", "equidistant", "geodesic") or a list of electrode positions (e.g. "Cz", "Pz").
cfg.eeg.EEGGround                     = ft_getopt(cfg.eeg, 'EEGGround'                   ); % RECOMMENDED. Description of the location of the ground electrode (e.g., "placed on right mastoid (M2)").
cfg.eeg.HardwareFilters               = ft_getopt(cfg.eeg, 'HardwareFilters'             ); % List of hardware (amplifier) filters applied or ideally  key:value pairs of pre-applied filters and their parameter values
cfg.eeg.SubjectArtefactDescription    = ft_getopt(cfg.eeg, 'SubjectArtefactDescription'  ); % Freeform description of the observed subject artefact and its possible cause (e.g. "Vagus Nerve Stimulator", "non-removable implant"). If this field is left empty, it will be interpreted as absence of  a source of (constantly present) artifacts.

%% iEEG specific fields
cfg.ieeg.iEEGReference                   = ft_getopt(cfg.ieeg, 'iEEGReference'                  ); % REQUIRED. General description of the reference scheme used and (when applicable) of location of the reference electrode in the raw recordings (e.g. "left mastoidâ€?, “bipolarâ€?, “T01â€? for electrode with name T01, “intracranial electrode on top of a grid, not included with dataâ€?, “upside down electrodeâ€?). If different channels have a different reference, this field should have a general description and the channel specific reference should be defined in the _channels.tsv file.
cfg.ieeg.SamplingFrequency               = ft_getopt(cfg.ieeg, 'SamplingFrequency'              ); % REQUIRED. Sampling frequency (in Hz) of all the iEEG channels in the recording (e.g., 2400). All other channels should have frequency specified as well in the channels.tsv file.
cfg.ieeg.PowerLineFrequency              = ft_getopt(cfg.ieeg, 'PowerLineFrequency'             ); % REQUIRED. Frequency (in Hz) of the power grid where the iEEG recording was done (i.e. 50 or 60)
cfg.ieeg.SoftwareFilters                 = ft_getopt(cfg.ieeg, 'SoftwareFilters'                ); % REQUIRED. List of temporal software filters applied or ideally  key:value pairs of pre-applied filters and their parameter values. (n/a if none).
% cfg.ieeg.DCOffsetCorrection see https://github.com/bids-standard/bids-specification/issues/237
cfg.ieeg.HardwareFilters                 = ft_getopt(cfg.ieeg, 'HardwareFilters'                ); % REQUIRED. List of hardware (amplifier) filters applied with  key:value pairs of filter parameters and their values.
cfg.ieeg.ElectrodeManufacturer           = ft_getopt(cfg.ieeg, 'ElectrodeManufacturer'          ); % RECOMMENDED. can be used if all electrodes are of the same manufacturer (e.g. AD-TECH, DIXI). If electrodes of different manufacturers are used, please use the corresponding table in the _electrodes.tsv file.
cfg.ieeg.ElectrodeManufacturersModelName = ft_getopt(cfg.ieeg, 'ElectrodeManufacturersModelName'); % RECOMMENDED. If different electrode types are used, please use the corresponding table in the _electrodes.tsv file.
% Manufacturer and ManufacturersModelName are general
cfg.ieeg.ECOGChannelCount                = ft_getopt(cfg.ieeg, 'ECOGChannelCount'               ); % RECOMMENDED. Number of iEEG surface channels included in the recording (e.g. 120)
cfg.ieeg.SEEGChannelCount                = ft_getopt(cfg.ieeg, 'SEEGChannelCount'               ); % RECOMMENDED. Number of iEEG depth channels included in the recording (e.g. 8)
cfg.ieeg.EEGChannelCount                 = ft_getopt(cfg.ieeg, 'EEGChannelCount'                ); % RECOMMENDED. Number of scalp EEG channels recorded simultaneously (e.g. 21)
cfg.ieeg.EOGChannelCount                 = ft_getopt(cfg.ieeg, 'EOGChannelCount'                ); % RECOMMENDED. Number of EOG channels
cfg.ieeg.ECGChannelCount                 = ft_getopt(cfg.ieeg, 'ECGChannelCount'                ); % RECOMMENDED. Number of ECG channels
cfg.ieeg.EMGChannelCount                 = ft_getopt(cfg.ieeg, 'EMGChannelCount'                ); % RECOMMENDED. Number of EMG channels
cfg.ieeg.MiscChannelCount                = ft_getopt(cfg.ieeg, 'MiscChannelCount'               ); % RECOMMENDED. Number of miscellaneous analog channels for auxiliary  signals
cfg.ieeg.TriggerChannelCount             = ft_getopt(cfg.ieeg, 'TriggerChannelCount'            ); % RECOMMENDED. Number of channels for digital (TTL bit level) triggers
cfg.ieeg.RecordingDuration               = ft_getopt(cfg.ieeg, 'RecordingDuration'              ); % RECOMMENDED. Length of the recording in seconds (e.g. 3600)
cfg.ieeg.RecordingType                   = ft_getopt(cfg.ieeg, 'RecordingType'                  ); % RECOMMENDED. Defines whether the recording is  “continuousâ€? or  “epochedâ€?; this latter limited to time windows about events of interest (e.g., stimulus presentations, subject responses etc.)
cfg.ieeg.EpochLength                     = ft_getopt(cfg.ieeg, 'EpochLength'                    ); % RECOMMENDED. Duration of individual epochs in seconds (e.g. 1) in case of epoched data
cfg.ieeg.iEEGGround                      = ft_getopt(cfg.ieeg, 'iEEGGround'                     ); % RECOMMENDED. Description  of the location of the ground electrode (“placed on right mastoid (M2)â€?).
cfg.ieeg.iEEGPlacementScheme             = ft_getopt(cfg.ieeg, 'iEEGPlacementScheme'            ); % RECOMMENDED. Freeform description of the placement of the iEEG electrodes. Left/right/bilateral/depth/surface (e.g. “left frontal grid and bilateral hippocampal depthâ€? or “surface strip and STN depthâ€? or “clinical indication bitemporal, bilateral temporal strips and left gridâ€?).
cfg.ieeg.iEEGElectrodeGroups             = ft_getopt(cfg.ieeg, 'iEEGElectrodeGroups'            ); % RECOMMENDED. Field to describe the way electrodes are grouped into strips, grids or depth probes e.g. {'grid1': "10x8 grid on left temporal pole", 'strip2': "1x8 electrode strip on xxx"}.
cfg.ieeg.SubjectArtefactDescription      = ft_getopt(cfg.ieeg, 'SubjectArtefactDescription'     ); % RECOMMENDED. Freeform description of the observed subject artefact and its possible cause (e.g. “door openâ€?, â€?nurse walked into room at 2 minâ€?, â€?seizure at 10 minâ€?). If this field is left empty, it will be interpreted as absence of artifacts.
cfg.ieeg.ElectricalStimulation           = ft_getopt(cfg.ieeg, 'ElectricalStimulation'          ); % OPTIONAL. Boolean field to specify if electrical stimulation was done during the recording (options are “trueâ€? or “falseâ€?). Parameters for event-like stimulation should be specified in the _events.tsv file (see example underneath).
cfg.ieeg.ElectricalStimulationParameters = ft_getopt(cfg.ieeg, 'ElectricalStimulationParameters'); % OPTIONAL. Free form description of stimulation parameters, such as frequency, shape etc. Specific onsets can be specified in the _events.tsv file. Specific shapes can be described here in freeform text.

%% EMG is not part of the official BIDS specification
cfg.emg.SamplingFrequency                 = ft_getopt(cfg.emg, 'SamplingFrequency'                 );
cfg.emg.RecordingDuration                 = ft_getopt(cfg.emg, 'RecordingDuration'                 );
cfg.emg.RecordingType                     = ft_getopt(cfg.emg, 'RecordingType'                     );
cfg.emg.PowerLineFrequency                = ft_getopt(cfg.emg, 'PowerLineFrequency'                );
cfg.emg.HardwareFilters                   = ft_getopt(cfg.emg, 'HardwareFilters'                   );
cfg.emg.SoftwareFilters                   = ft_getopt(cfg.emg, 'SoftwareFilters'                   );
cfg.emg.EMGChannelCount                   = ft_getopt(cfg.emg, 'EMGChannelCount'                   );
cfg.emg.EOGChannelCount                   = ft_getopt(cfg.emg, 'EOGChannelCount'                   );
cfg.emg.ECGChannelCount                   = ft_getopt(cfg.emg, 'ECGChannelCount'                   );
cfg.emg.ElectrodeManufacturer             = ft_getopt(cfg.emg, 'ElectrodeManufacturer'             );
cfg.emg.ElectrodeManufacturersModelName   = ft_getopt(cfg.emg, 'ElectrodeManufacturersModelName'   );
cfg.emg.EMGPlacementScheme                = ft_getopt(cfg.emg, 'EMGPlacementScheme'                );
cfg.emg.EMGReference                      = ft_getopt(cfg.emg, 'EMGReference'                      );
cfg.emg.EMGGround                         = ft_getopt(cfg.emg, 'EMGGround'                         );

%% EXG is not part of the official BIDS specification
cfg.exg.SamplingFrequency                 = ft_getopt(cfg.exg, 'SamplingFrequency'                 );
cfg.exg.RecordingDuration                 = ft_getopt(cfg.exg, 'RecordingDuration'                 );
cfg.exg.RecordingType                     = ft_getopt(cfg.exg, 'RecordingType'                     );

%% NIRS specific fields
cfg.nirs.CapManufacturer                   = ft_getopt(cfg.nirs, 'CapManufacturer'                   );
cfg.nirs.CapManufacturersModelName         = ft_getopt(cfg.nirs, 'CapManufacturersModelName'         );
cfg.nirs.SamplingFrequency                 = ft_getopt(cfg.nirs, 'SamplingFrequency'                 );
cfg.nirs.NIRSChannelCount                  = ft_getopt(cfg.nirs, 'NIRSChannelCount'                  );
cfg.nirs.NIRSSourceOptodeCount             = ft_getopt(cfg.nirs, 'NIRSSourceOptodeCount'             );
cfg.nirs.NIRSDetectorOptodeCount           = ft_getopt(cfg.nirs, 'NIRSDetectorOptodeCount'           );
cfg.nirs.ACCELChannelCount                 = ft_getopt(cfg.nirs, 'ACCELChannelCount'                 );
cfg.nirs.GYROChannelCount                  = ft_getopt(cfg.nirs, 'GYROChannelCount'                  );
cfg.nirs.MAGNChannelCount                  = ft_getopt(cfg.nirs, 'MAGNChannelCount'                  );
cfg.nirs.SourceType                        = ft_getopt(cfg.nirs, 'SourceType'                        );
cfg.nirs.DetectorType                      = ft_getopt(cfg.nirs, 'DetectorType'                      );
cfg.nirs.ShortChannelCount                 = ft_getopt(cfg.nirs, 'ShortChannelCount'                 );
cfg.nirs.NIRSPlacementScheme               = ft_getopt(cfg.nirs, 'NIRSPlacementScheme'               );
cfg.nirs.RecordingDuration                 = ft_getopt(cfg.nirs, 'RecordingDuration'                 );
cfg.nirs.DCOffsetCorrection                = ft_getopt(cfg.nirs, 'DCOffsetCorrection'                );
cfg.nirs.HeadCircumference                 = ft_getopt(cfg.nirs, 'HeadCircumference'                 );
cfg.nirs.HardwareFilters                   = ft_getopt(cfg.nirs, 'HardwareFilters'                   );
cfg.nirs.SoftwareFilters                   = ft_getopt(cfg.nirs, 'SoftwareFilters'                   );
cfg.nirs.SubjectArtefactDescription        = ft_getopt(cfg.nirs, 'SubjectArtefactDescription'        );

%% audio is not part of the official BIDS specification
cfg.audio.SampleRate                      = ft_getopt(cfg.audio, 'SampleRate'        );
cfg.audio.ChannelCount                    = ft_getopt(cfg.audio, 'ChannelCount'      );
cfg.audio.RecordingDuration               = ft_getopt(cfg.audio, 'RecordingDuration' );

%% video is not part of the official BIDS specification
cfg.video.FrameRate                       = ft_getopt(cfg.video, 'FrameRate'           );
cfg.video.Width                           = ft_getopt(cfg.video, 'Width'               );
cfg.video.Height                          = ft_getopt(cfg.video, 'Height'              );
cfg.video.BitsPerPixel                    = ft_getopt(cfg.video, 'BitsPerPixel'        );
cfg.video.VideoDuration                   = ft_getopt(cfg.video, 'VideoDuration'       );
cfg.video.AudioDuration                   = ft_getopt(cfg.video, 'AudioDuration'       );
cfg.video.AudioSampleRate                 = ft_getopt(cfg.video, 'AudioSampleRate'     );
cfg.video.AudioChannelCount               = ft_getopt(cfg.video, 'AudioChannelCount'   );

%% physio is part of the official BIDS specification and goes into 'beh' or in 'func'
cfg.physio.Columns                        = ft_getopt(cfg.physio, 'Columns'            );
cfg.physio.StartTime                      = ft_getopt(cfg.physio, 'StartTime'          );
cfg.physio.SamplingFrequency              = ft_getopt(cfg.physio, 'SamplingFrequency'  );

%% stim is part of the official BIDS specification and goes into 'beh' or in 'func'
cfg.stim.Columns                          = ft_getopt(cfg.stim, 'Columns'              );
cfg.stim.StartTime                        = ft_getopt(cfg.stim, 'StartTime'            );
cfg.stim.SamplingFrequency                = ft_getopt(cfg.stim, 'SamplingFrequency'    );

%% eyetracker is not part of the official BIDS specification
% this follows https://bids-specification.readthedocs.io/en/stable/04-modality-specific-files/06-physiological-and-other-continuous-recordings.html
cfg.eyetracker.Columns                    = ft_getopt(cfg.eyetracker, 'Columns'              );
cfg.eyetracker.StartTime                  = ft_getopt(cfg.eyetracker, 'StartTime'            );
cfg.eyetracker.SamplingFrequency          = ft_getopt(cfg.eyetracker, 'SamplingFrequency'    );

%% motion is not part of the official BIDS specification
% this follows extension proposal 029 https://bids.neuroimaging.io/bep029
cfg.motion.EpochLength                    = ft_getopt(cfg.motion, 'EpochLength'               );
cfg.motion.RecordingType                  = ft_getopt(cfg.motion, 'RecordingType'             );
cfg.motion.SubjectArtefactDescription     = ft_getopt(cfg.motion, 'SubjectArtefactDescription');
cfg.motion.TrackingSystemName             = ft_getopt(cfg.motion, 'TrackingSystemName'        );
cfg.motion.DeviceSerialNumber             = ft_getopt(cfg.motion, 'DeviceSerialNumber'        );
cfg.motion.SoftwareVersions               = ft_getopt(cfg.motion, 'SoftwareVersions'          );
cfg.motion.ExternalSoftwareVersions       = ft_getopt(cfg.motion, 'ExternalSoftwareVersions'  );
cfg.motion.Manufacturer                   = ft_getopt(cfg.motion, 'Manufacturer'              );
cfg.motion.ManufacturersModelName         = ft_getopt(cfg.motion, 'ManufacturersModelName'    );
cfg.motion.SamplingFrequency              = ft_getopt(cfg.motion, 'SamplingFrequency'         );
cfg.motion.SamplingFrequencyEffective     = ft_getopt(cfg.motion, 'SamplingFrequencyEffective');
cfg.motion.RecordingDuration              = ft_getopt(cfg.motion, 'RecordingDuration'         );
cfg.motion.SpatialAxes                    = ft_getopt(cfg.motion, 'SpatialAxes'               );
cfg.motion.RotationRule                   = ft_getopt(cfg.motion, 'RotationRule'              );
cfg.motion.RotationOrder                  = ft_getopt(cfg.motion, 'RotationOrder'             );

%% information for the coordsystem.json file for MEG, EEG, iEEG, and motion
cfg.coordsystem.MEGCoordinateSystem                             = ft_getopt(cfg.coordsystem, 'MEGCoordinateSystem'                            ); % REQUIRED. Defines the coordinate system for the MEG sensors. See Appendix VIII: preferred names of Coordinate systems. If "Other", provide definition of the coordinate system in [MEGCoordinateSystemDescription].
cfg.coordsystem.MEGCoordinateUnits                              = ft_getopt(cfg.coordsystem, 'MEGCoordinateUnits'                             ); % REQUIRED. Units of the coordinates of MEGCoordinateSystem. MUST be ???m???, ???cm???, or ???mm???.
cfg.coordsystem.MEGCoordinateSystemDescription                  = ft_getopt(cfg.coordsystem, 'MEGCoordinateSystemDescription'                 ); % OPTIONAL. Freeform text description or link to document describing the MEG coordinate system system in detail.
cfg.coordsystem.HeadCoilCoordinates                             = ft_getopt(cfg.coordsystem, 'HeadCoilCoordinates'                     ); % OPTIONAL. Key:value pairs describing head localization coil labels and their coordinates, interpreted following the HeadCoilCoordinateSystem, e.g., {"NAS": [12.7,21.3,13.9], "LPA": [5.2,11.3,9.6], "RPA": [20.2,11.3,9.1]}. Note that coils are not always placed at locations that have a known anatomical name (e.g. for Neuromag/Elekta, Yokogawa systems); in that case generic labels can be used (e.g. {"coil1": [122,213,123], "coil2": [67,123,86], "coil3": [219,110,81]} ).
cfg.coordsystem.HeadCoilCoordinateSystem                        = ft_getopt(cfg.coordsystem, 'HeadCoilCoordinateSystem'                           ); % OPTIONAL. Defines the coordinate system for the coils. See Appendix VIII: preferred names of Coordinate systems. If "Other", provide definition of the coordinate system in HeadCoilCoordinateSystemDescription.
cfg.coordsystem.HeadCoilCoordinateUnits                         = ft_getopt(cfg.coordsystem, 'HeadCoilCoordinateUnits'                        ); % OPTIONAL. Units of the coordinates of HeadCoilCoordinateSystem. MUST be ???m???, ???cm???, or ???mm???.
cfg.coordsystem.HeadCoilCoordinateSystemDescription             = ft_getopt(cfg.coordsystem, 'HeadCoilCoordinateSystemDescription'            ); % OPTIONAL. Freeform text description or link to document describing the Head Coil coordinate system system in detail.
cfg.coordsystem.DigitizedHeadPoints                             = ft_getopt(cfg.coordsystem, 'DigitizedHeadPoints'                            ); % OPTIONAL. Relative path to the file containing the locations of digitized head points collected during the session (e.g., "sub-01_headshape.pos"). RECOMMENDED for all MEG systems, especially for CTF and 4D/BTi. For Neuromag/Elekta/Megin the head points will be stored in the fif file.
cfg.coordsystem.DigitizedHeadPointsCoordinateSystem             = ft_getopt(cfg.coordsystem, 'DigitizedHeadPointsCoordinateSystem'            ); % OPTIONAL. Defines the coordinate system for the digitized head points. See Appendix VIII: preferred names of Coordinate systems. If "Other", provide definition of the coordinate system in DigitizedHeadPointsCoordinateSystemDescription.
cfg.coordsystem.DigitizedHeadPointsCoordinateUnits              = ft_getopt(cfg.coordsystem, 'DigitizedHeadPointsCoordinateUnits'             ); % OPTIONAL. Units of the coordinates of DigitizedHeadPointsCoordinateSystem. MUST be ???m???, ???cm???, or ???mm???.
cfg.coordsystem.DigitizedHeadPointsCoordinateSystemDescription  = ft_getopt(cfg.coordsystem, 'DigitizedHeadPointsCoordinateSystemDescription' ); % OPTIONAL. Freeform text description or link to document describing the Digitized head Points coordinate system system in detail.
cfg.coordsystem.EEGCoordinateSystem                             = ft_getopt(cfg.coordsystem, 'EEGCoordinateSystem'                            ); % OPTIONAL. Describes how the coordinates of the EEG sensors are to be interpreted.
cfg.coordsystem.EEGCoordinateUnits                              = ft_getopt(cfg.coordsystem, 'EEGCoordinateUnits'                             ); % OPTIONAL. Units of the coordinates of EEGCoordinateSystem. MUST be ???m???, ???cm???, or ???mm???.
cfg.coordsystem.EEGCoordinateSystemDescription                  = ft_getopt(cfg.coordsystem, 'EEGCoordinateSystemDescription'                 ); % OPTIONAL. Freeform text description or link to document describing the EEG coordinate system system in detail.
cfg.coordsystem.iEEGCoordinateSystem                            = ft_getopt(cfg.coordsystem, 'iEEGCoordinateSystem'                           ); % REQUIRED. Defines the coordinate system for the iEEG electrodes. See Appendix VIII for a list of restricted keywords. If positions correspond to pixel indices in a 2D image (of either a volume-rendering, surface-rendering, operative photo, or operative drawing), this must be "Pixels". For more information, see the section on 2D coordinate systems
cfg.coordsystem.iEEGCoordinateUnits	                            = ft_getopt(cfg.coordsystem, 'iEEGCoordinateUnits'                            ); % REQUIRED. Units of the _electrodes.tsv, MUST be "m", "mm", "cm" or "pixels".
cfg.coordsystem.iEEGCoordinateSystemDescription	                = ft_getopt(cfg.coordsystem, 'iEEGCoordinateSystemDescription'                ); % RECOMMENDED. Freeform text description or link to document describing the iEEG coordinate system system in detail (e.g., "Coordinate system with the origin at anterior commissure (AC), negative y-axis going through the posterior commissure (PC), z-axis going to a mid-hemisperic point which lies superior to the AC-PC line, x-axis going to the right").
cfg.coordsystem.iEEGCoordinateProcessingDescription             = ft_getopt(cfg.coordsystem, 'iEEGCoordinateProcessingDescription'            ); % RECOMMENDED. Has any post-processing (such as projection) been done on the electrode positions (e.g., "surface_projection", "none").
cfg.coordsystem.iEEGCoordinateProcessingReference	              = ft_getopt(cfg.coordsystem, 'iEEGCoordinateProcessingReference'              ); % RECOMMENDED. A reference to a paper that defines in more detail the method used to localize the electrodes and to post-process the electrode positions. .
cfg.coordsystem.NIRSCoordinateSystem                            = ft_getopt(cfg.coordsystem, 'NIRSCoordinateSystem'                           ); % REQUIRED. Defines the coordinate system for the optodes. See Appendix VIII for a list of restricted keywords. If positions correspond to pixel indices in a 2D image (of either a volume-rendering, surface-rendering, operative photo, or operative drawing), this must be "Pixels". For more information, see the section on 2D coordinate systems
cfg.coordsystem.NIRSCoordinateUnits                             = ft_getopt(cfg.coordsystem, 'NIRSCoordinateUnits'                            ); % REQUIRED. Units of the _optodes.tsv, MUST be "m", "mm", "cm" or "pixels".
cfg.coordsystem.NIRSCoordinateSystemDescription                 = ft_getopt(cfg.coordsystem, 'NIRSCoordinateSystemDescription'                ); % RECOMMENDED. Freeform text description or link to document describing the NIRS coordinate system system in detail (e.g., "Coordinate system with the origin at anterior commissure (AC), negative y-axis going through the posterior commissure (PC), z-axis going to a mid-hemisperic point which lies superior to the AC-PC line, x-axis going to the right").
cfg.coordsystem.NIRSCoordinateProcessingDescription             = ft_getopt(cfg.coordsystem, 'NIRSCoordinateProcessingDescription'            ); % RECOMMENDED. Has any post-processing (such as projection) been done on the optode positions (e.g., "surface_projection", "none").
cfg.coordsystem.MotionCoordinateSystem              	          = ft_getopt(cfg.coordsystem, 'MotionCoordinateSystem'                         ); % REQUIRED.
cfg.coordsystem.MotionRotationRule              	              = ft_getopt(cfg.coordsystem, 'MotionRotationRule'                             ); % OPTIONAL.
cfg.coordsystem.MotionRotationOrder              	              = ft_getopt(cfg.coordsystem, 'MotionRotationOrder'                            ); % OPTIONAL.
cfg.coordsystem.IntendedFor                                     = ft_getopt(cfg.coordsystem, 'IntendedFor'                                    ); % OPTIONAL. Path or list of path relative to the subject subfolder pointing to the structural MRI, possibly of different types if a list is specified, to be used with the MEG recording. The path(s) need(s) to use forward slashes instead of backward slashes (e.g. "ses-<label>/anat/sub-01_T1w.nii.gz").
cfg.coordsystem.AnatomicalLandmarkCoordinates                   = ft_getopt(cfg.coordsystem, 'AnatomicalLandmarkCoordinates'                  ); % OPTIONAL. Key:value pairs of the labels and 3-D digitized locations of anatomical landmarks, interpreted following the AnatomicalLandmarkCoordinateSystem, e.g., {"NAS": [12.7,21.3,13.9], "LPA": [5.2,11.3,9.6], "RPA": [20.2,11.3,9.1]}.
cfg.coordsystem.AnatomicalLandmarkCoordinateSystem              = ft_getopt(cfg.coordsystem, 'AnatomicalLandmarkCoordinateSystem'             ); % OPTIONAL. Defines the coordinate system for the anatomical landmarks. See Appendix VIII: preferred names of Coordinate systems. If "Other", provide definition of the coordinate system in AnatomicalLandmarkCoordinateSystemDescripti on.
cfg.coordsystem.AnatomicalLandmarkCoordinateUnits               = ft_getopt(cfg.coordsystem, 'AnatomicalLandmarkCoordinateUnits'              ); % OPTIONAL. Units of the coordinates of AnatomicalLandmarkCoordinateSystem. MUST be ???m???, ???cm???, or ???mm???.
cfg.coordsystem.AnatomicalLandmarkCoordinateSystemDescription   = ft_getopt(cfg.coordsystem, 'AnatomicalLandmarkCoordinateSystemDescription'  ); % OPTIONAL. Freeform text description or link to document describing the Head Coil coordinate system system in detail.
cfg.coordsystem.FiducialsDescription                            = ft_getopt(cfg.coordsystem, 'FiducialsDescription'                           ); % OPTIONAL. A freeform text field documenting the anatomical landmarks that were used and how the head localization coils were placed relative to these. This field can describe, for instance, whether the true anatomical locations of the left and right pre-auricular points were used and digitized, or rather whether they were defined as the intersection between the tragus and the helix (the entry of the ear canal), or any other anatomical description of selected points in the vicinity of the ears.
cfg.coordsystem.FiducialsCoordinates                            = ft_getopt(cfg.coordsystem, 'FiducialsCoordinates'                           ); % RECOMMENDED. Key:value pairs of the labels and 3-D digitized position of anatomical landmarks, interpreted following the FiducialsCoordinateSystem (e.g., {"NAS": [12.7,21.3,13.9], "LPA": [5.2,11.3,9.6], "RPA": [20.2,11.3,9.1]}).
cfg.coordsystem.FiducialsCoordinateSystem                       = ft_getopt(cfg.coordsystem, 'FiducialsCoordinateSystem'                      ); % RECOMMENDED. Refers to the coordinate space to which the landmarks positions are to be interpreted - preferably the same as the NIRSCoordinateSystem
cfg.coordsystem.FiducialsCoordinateUnits                        = ft_getopt(cfg.coordsystem, 'FiducialsCoordinateUnits'                       ); % RECOMMENDED. Units in which the coordinates that are listed in the field AnatomicalLandmarkCoordinateSystem are represented (e.g., "mm", "cm").
cfg.coordsystem.FiducialsCoordinateSystemDescription            = ft_getopt(cfg.coordsystem, 'FiducialsCoordinateSystemDescription'           ); % RECOMMENDED. Free-form text description of the coordinate system. May also include a link to a documentation page or paper describing the system in greater detail.

%% columns in the channels.tsv
cfg.channels.name                       = ft_getopt(cfg.channels, 'name'                        , nan);  % REQUIRED. Channel name (e.g., MRT012, MEG023)
cfg.channels.type                       = ft_getopt(cfg.channels, 'type'                        , nan);  % REQUIRED. Type of channel; MUST use the channel types listed below.
cfg.channels.units                      = ft_getopt(cfg.channels, 'units'                       , nan);  % REQUIRED. Physical unit of the data values recorded by this channel in SI (see Appendix V: Units for allowed symbols).
% specific options for EEG/MEG/iEEG channels
cfg.channels.sampling_frequency         = ft_getopt(cfg.channels, 'sampling_frequency'          , nan);  % OPTIONAL. Sampling rate of the channel in Hz.
cfg.channels.description                = ft_getopt(cfg.channels, 'description'                 , nan);  % OPTIONAL. Brief free-text description of the channel, or other information of interest. See examples below.
cfg.channels.low_cutoff                 = ft_getopt(cfg.channels, 'low_cutoff'                  , nan);  % OPTIONAL. Frequencies used for the high-pass filter applied to the channel in Hz. If no high-pass filter applied, use n/a.
cfg.channels.high_cutoff                = ft_getopt(cfg.channels, 'high_cutoff'                 , nan);  % OPTIONAL. Frequencies used for the low-pass filter applied to the channel in Hz. If no low-pass filter applied, use n/a. Note that hardware anti-aliasing in A/D conversion of all MEG/EEG electronics applies a low-pass filter; specify its frequency here if applicable.
cfg.channels.notch                      = ft_getopt(cfg.channels, 'notch'                       , nan);  % OPTIONAL. Frequencies used for the notch filter applied to the channel, in Hz. If no notch filter applied, use n/a.
cfg.channels.software_filters           = ft_getopt(cfg.channels, 'software_filters'            , nan);  % OPTIONAL. List of temporal and/or spatial software filters applied (e.g. "SSS", "SpatialCompensation"). Note that parameters should be defined in the general MEG sidecar .json file. Indicate n/a in the absence of software filters applied.
cfg.channels.status                     = ft_getopt(cfg.channels, 'status'                      , nan);  % OPTIONAL. Data quality observed on the channel (good/bad). A channel is considered bad if its data quality is compromised by excessive noise. Description of noise type SHOULD be provided in [status_description].
cfg.channels.status_description         = ft_getopt(cfg.channels, 'status_description'          , nan);  % OPTIONAL. Freeform text description of noise or artifact affecting data quality on the channel. It is meant to explain why the channel was declared bad in [status].
% specific options for NIRS channels
cfg.channels.source                     = ft_getopt(cfg.channels, 'source'                      , nan);
cfg.channels.detector                   = ft_getopt(cfg.channels, 'detector'                    , nan);
cfg.channels.wavelength_nominal         = ft_getopt(cfg.channels, 'wavelength_nominal'          , nan);
cfg.channels.wavelength_actual          = ft_getopt(cfg.channels, 'wavelength_actual'           , nan);
cfg.channels.wavelength_emission_actual = ft_getopt(cfg.channels, 'wavelength_emission_actual'  , nan);
cfg.channels.short_channel              = ft_getopt(cfg.channels, 'short_channel'               , nan);
% specific options for motion channels
cfg.channels.sampling_frequency         = ft_getopt(cfg.channels, 'sampling_frequency'          , nan);
cfg.channels.orientation_component      = ft_getopt(cfg.channels, 'component'                   , nan);
cfg.channels.tracked_point              = ft_getopt(cfg.channels, 'tracked_point'               , nan);

%% columns in the electrodes.tsv
cfg.electrodes.name             = ft_getopt(cfg.electrodes, 'name'             , nan);  % REQUIRED. Name of the electrode
cfg.electrodes.x                = ft_getopt(cfg.electrodes, 'x'                , nan);  % REQUIRED. Recorded position along the x-axis
cfg.electrodes.y                = ft_getopt(cfg.electrodes, 'y'                , nan);  % REQUIRED. Recorded position along the y-axis
cfg.electrodes.z                = ft_getopt(cfg.electrodes, 'z'                , nan);  % REQUIRED. Recorded position along the z-axis
cfg.electrodes.type             = ft_getopt(cfg.electrodes, 'type'             , nan);  % RECOMMENDED. Type of the electrode (e.g., cup, ring, clip-on, wire, needle)
cfg.electrodes.material         = ft_getopt(cfg.electrodes, 'material'         , nan);  % RECOMMENDED. Material of the electrode, e.g., Tin, Ag/AgCl, Gold
cfg.electrodes.impedance        = ft_getopt(cfg.electrodes, 'impedance'        , nan);  % RECOMMENDED. Impedance of the electrode in kOhm

%% columns in the optodes.tsv
cfg.optodes.name                = ft_getopt(cfg.optodes, 'name'                , nan);  % REQUIRED. Name of the optode must be unique
cfg.optodes.type                = ft_getopt(cfg.optodes, 'type'                , nan);  % REQUIRED. Either source or detector
cfg.optodes.x                   = ft_getopt(cfg.optodes, 'x'                   , nan);  % REQUIRED. Recorded position along the x-axis. n/a if not available
cfg.optodes.y                   = ft_getopt(cfg.optodes, 'y'                   , nan);  % REQUIRED. Recorded position along the y-axis. n/a if not available
cfg.optodes.z                   = ft_getopt(cfg.optodes, 'z'                   , nan);  % REQUIRED. Recorded position along the z-axis. n/a if not available
cfg.optodes.template_x          = ft_getopt(cfg.optodes, 'template_x'          , nan);  % OPTIONAL. Assumed or ideal position along the x axis
cfg.optodes.template_y          = ft_getopt(cfg.optodes, 'template_y'          , nan);  % OPTIONAL. Assumed or ideal position along the x axis
cfg.optodes.template_z          = ft_getopt(cfg.optodes, 'template_z'          , nan);  % OPTIONAL. Assumed or ideal position along the x axis
cfg.optodes.description         = ft_getopt(cfg.optodes, 'description'         , nan);  % OPTIONAL.	string	Free-form text description of the optode, or other information of interest.
cfg.optodes.detector_type       = ft_getopt(cfg.optodes, 'detector_type'       , nan);  % OPTIONAL.	string	The type of detector. Only to be used if the field DetectorType in *_nirs.json is set to mixed.
cfg.optodes.source_type         = ft_getopt(cfg.optodes, 'source_type'         , nan);  % OPTIONAL.	string	The type of source. Only to be used if the field SourceType in *_nirs.json is set to mixed.


%% information for the participants.tsv
cfg.participants = ft_getopt(cfg, 'participants', struct());

%% information for the scans.tsv
cfg.scans = ft_getopt(cfg, 'scans', struct());

%% information for the sessions.tsv
cfg.sessions = ft_getopt(cfg, 'sessions', struct());

%% sanity checks and determine the default method/outputfile

% the task is both part of the file name (cfg.task) and is also one of the general JSON metadata fields (cfg.TaskName)
if isempty(cfg.task) && isempty(cfg.TaskName)
    % this is fine
elseif ~isempty(cfg.task) && isempty(cfg.TaskName)
    cfg.TaskName = cfg.task;
elseif isempty(cfg.task) && ~isempty(cfg.TaskName)
    cfg.task = cfg.TaskName;
elseif ~isempty(cfg.task) && ~isempty(cfg.TaskName)
    if ~strcmp(cfg.task, cfg.TaskName)
        ft_error('cfg.task and cfg.TaskName should be identical');
    end
end

% construct the output directory and file name
if isempty(cfg.outputfile)
    if isempty(cfg.method) && isempty(cfg.bidsroot) && ~isempty(cfg.dataset)
        cfg.outputfile = cfg.dataset;
        cfg.method = 'decorate';
        ft_notice('using cfg.outputfile=''%s'' and cfg.method=''%s''', cfg.outputfile, cfg.method);
    elseif isempty(cfg.bidsroot)
        ft_error('cfg.bidsroot is required to construct BIDS output directory and file');
    elseif isempty(cfg.sub)
        ft_error('cfg.sub is required to construct BIDS output directory and file');
    elseif isempty(cfg.datatype)
        ft_error('cfg.datatype is required to construct BIDS output directory and file');
    else
        dirname = datatype2dirname(cfg.datatype);
        filename = ['sub-' cfg.sub];
        filename = add_entity(filename, 'ses',  cfg.ses);
        filename = add_entity(filename, 'task', cfg.task);
        filename = add_entity(filename, 'tracksys', cfg.tracksys);
        filename = add_entity(filename, 'acq',  cfg.acq);
        filename = add_entity(filename, 'ce',   cfg.ce);
        filename = add_entity(filename, 'rec',  cfg.rec);
        filename = add_entity(filename, 'dir',  cfg.dir);
        filename = add_entity(filename, 'run',  cfg.run);
        filename = add_entity(filename, 'mod',  cfg.mod);
        filename = add_entity(filename, 'echo', cfg.echo);
        filename = add_entity(filename, 'proc', cfg.proc);
        filename = add_entity(filename, 'desc', cfg.desc);
        filename = add_datatype(filename, cfg.datatype);
        if ~isempty(cfg.ses)
            % construct the output filename, with session directory
            cfg.outputfile = fullfile(cfg.bidsroot, ['sub-' cfg.sub], ['ses-' cfg.ses], dirname, filename);
        else
            % construct the output filename, without session directory
            cfg.outputfile = fullfile(cfg.bidsroot, ['sub-' cfg.sub], dirname, filename);
        end
        if strcmp(cfg.method, 'copy') && ~isempty(cfg.dataset)
            % copy the file extension from the input dataset
            [p, f, x] = fileparts(cfg.dataset);
            cfg.outputfile = [cfg.outputfile x];
        end
    end
end

% set the default method
if isempty(cfg.method)
    if ~isempty(cfg.dataset) && ~isequal(cfg.dataset, cfg.outputfile)
        cfg.method = 'convert';
    elseif isempty(cfg.dataset) && ~isempty(varargin)
        cfg.method = 'convert';
    else
        cfg.method = 'decorate';
    end
    ft_notice('using cfg.method=''%s''', cfg.method);
end

% do some sanity checks on the input and the method
if istrue(cfg.deface) && ~strcmp(cfg.method, 'convert')
    ft_error('defacing only works in combination with cfg.method=''convert''');
elseif ft_nargin>1 && ~strcmp(cfg.method, 'convert')
    ft_error('input data only works in combination with cfg.method=''convert''');
end

% do some more sanity checks on the input and the method
switch cfg.method
    case 'decorate'
        if ~isempty(cfg.outputfile) && ~isempty(cfg.dataset) && ~isequal(cfg.dataset, cfg.outputfile)
            ft_error('cfg.dataset and cfg.outputfile should be the same');
        end
    case 'convert'
        if ~isempty(cfg.outputfile) && isequal(cfg.dataset, cfg.outputfile)
            ft_error('cfg.dataset and cfg.outputfile should not be the same');
        end
    case 'copy'
        if ~isempty(cfg.outputfile) && isequal(cfg.dataset, cfg.outputfile)
            ft_error('cfg.dataset and cfg.outputfile should not be the same');
        end
    otherwise
        ft_error('unsupported value for cfg.method')
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% read the information from the dataset
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% convert dataset to headerfile and datafile
if nargin>1
    % input data was specified
    varargin{1} = ft_checkdata(varargin{1}, 'datatype', {'raw', 'volume'});
    typ = ft_datatype(varargin{1});
elseif ~isempty(cfg.dataset)
    % data should be read from disk
    cfg = ft_checkconfig(cfg, 'dataset2files', 'yes');
    if isfield(cfg, 'headerfile')
        typ = ft_filetype(cfg.headerfile);
    else
        typ = ft_filetype(cfg.presentationfile);
    end
else
    typ = 'unknown';
end

% determine the json files that are required
need_mri_json           = false;
need_meg_json           = false;
need_eeg_json           = false;
need_ieeg_json          = false;
need_emg_json           = false;
need_exg_json           = false;
need_nirs_json          = false;
need_audio_json         = false;
need_video_json         = false;
need_physio_json        = false;
need_stim_json          = false;
need_eyetracker_json    = false;
need_motion_json        = false;
need_coordsystem_json   = false;
% determine the tsv files that are required
need_events_tsv         = false; % for functional and behavioral experiments
need_channels_tsv       = false; % only needed for MEG/EEG/iEEG/EMG/NIRS/motion
need_electrodes_tsv     = false; % only needed when actually present as cfg.electrodes, data.elec or as cfg.elec
need_optodes_tsv        = false; % only needed when actually present as cfg.optodes, data.opto or as cfg.opto

switch typ
    case {'nifti', 'nifti2', 'nifti_gz'}
        mri = ft_read_mri(cfg.dataset);
        if ~isempty(cfg.dicomfile)
            % read the header details from the matching DICOM file specified by the user
            dcm = dicominfo(cfg.dicomfile);
        else
            dcm = [];
        end
        need_mri_json = true;

    case 'dicom'
        mri = ft_read_mri(cfg.dataset);
        dcm = dicominfo(cfg.dataset);
        need_mri_json = true;

    case 'volume'
        % the input data structure represents imaging data
        mri = varargin{1};
        if ~isempty(cfg.dicomfile)
            % read the header details from the dicom matching file that was specified by the user
            dcm = dicominfo(cfg.dicomfile);
        elseif isfield(mri, 'hdr') && numel(mri.hdr)>1
            % it looks like an MRI read in using FT_READ_MRI using the FreeSurfer code
            % take the DICOM details from the first slice
            dcm = mri.hdr(1);
        else
            dcm = [];
        end
        need_mri_json = true;

    case {'ctf_ds', 'ctf_meg4', 'ctf_res4', 'ctf151', 'ctf275', 'neuromag_fif', 'neuromag122', 'neuromag306'}
        % it is MEG data from disk and in a supported format
        hdr = ft_read_header(cfg.headerfile, 'checkmaxfilter', false, 'readbids', false);
        if strcmp(cfg.method, 'convert')
            % the data should be converted and written to disk
            dat = ft_read_data(cfg.datafile, 'header', hdr, 'checkboundary', false, 'begsample', 1, 'endsample', hdr.nSamples*hdr.nTrials);
        end
        % read the triggers from disk
        trigger = ft_read_event(cfg.datafile, 'header', hdr, 'readbids', false);
        need_meg_json = true;

    case {'brainvision_vhdr', 'edf', 'eeglab_set', 'biosemi_bdf'}
        % the file on disk contains ExG data in a BIDS compiant format
        hdr = ft_read_header(cfg.headerfile, 'checkmaxfilter', false, 'readbids', false);
        if strcmp(cfg.method, 'convert')
            % the data should be converted and written to disk
            dat = ft_read_data(cfg.datafile, 'header', hdr, 'checkboundary', false, 'begsample', 1, 'endsample', hdr.nSamples*hdr.nTrials);
        end
        % read the triggers from disk
        trigger = ft_read_event(cfg.datafile, 'header', hdr, 'readbids', false);
        if isequal(cfg.datatype, 'eeg')
            need_eeg_json = true;
        elseif isequal(cfg.datatype, 'ieeg')
            need_ieeg_json = true;
        elseif isequal(cfg.datatype, 'emg')
            need_emg_json = true;
        elseif isequal(cfg.datatype, 'exg')
            need_exg_json = true;
        else
            ft_warning('assuming that the dataset represents EEG');
            need_eeg_json = true;
        end

    case 'presentation_log'
        trigger = ft_read_event(cfg.dataset);
        need_events_tsv = true;

    case {'audio_wav', 'audio_ogg', 'audio_flac', 'audio_au', 'audio_aiff', 'audio_aif', 'audio_aifc', 'audio_mp3', 'audio_m4a', 'audio_mp4'}
        % the file on disk contains audio
        need_audio_json = true;
        try
            audio = audioinfo(cfg.dataset);
        catch
            ft_warning('audio format is unsupported on this MATLAB version and/or operating system');
            audio = struct('SampleRate', nan, 'Duration', nan, 'NumChannels', nan);
        end

    case 'video'
        % the file on disk contains not only video, but also audio
        need_video_json = true;
        try
            video = VideoReader(cfg.dataset);
        catch
            ft_warning('video format is unsupported on this MATLAB version and/or operating system');
            video = struct('FrameRate', nan, 'Width', nan, 'Height', nan, 'Duration', nan);
        end
        try
            audio = audioinfo(cfg.dataset);
        catch
            ft_warning('audio format is unsupported on this MATLAB version and/or operating system');
            audio = struct('SampleRate', nan, 'Duration', nan, 'NumChannels', nan);
        end

    case 'raw'
        % the input data structure contains raw timeseries data
        if isequal(cfg.datatype, 'meg')
            need_meg_json = true;
        elseif isequal(cfg.datatype, 'eeg')
            need_eeg_json = true;
        elseif isequal(cfg.datatype, 'ieeg')
            need_ieeg_json = true;
        elseif isequal(cfg.datatype, 'emg')
            need_emg_json = true;
        elseif isequal(cfg.datatype, 'exg')
            need_exg_json = true;
        elseif isequal(cfg.datatype, 'nirs')
            need_nirs_json = true;
        elseif isequal(cfg.datatype, 'physio')
            need_physio_json = true;
        elseif isequal(cfg.datatype, 'stim')
            need_stim_json = true;
        elseif isequal(cfg.datatype, 'eyetracker')
            need_eyetracker_json = true;
        elseif isequal(cfg.datatype, 'motion')
            need_motion_json = true;
        else
            ft_error('cannot determine the type of the data, please specify cfg.dataype');
        end

        hdr = ft_fetch_header(varargin{1});
        if strcmp(cfg.method, 'convert')
            % the data should be written to disk
            dat     = ft_fetch_data(varargin{1}, 'checkboundary', false, 'begsample', 1, 'endsample', hdr.nSamples*hdr.nTrials);
            % the events should be writen to disk
            trigger = ft_fetch_event(varargin{1});
        end

        if ft_senstype(varargin{1}, 'ctf') || ft_senstype(varargin{1}, 'neuromag')
            % use the subsequent MEG-specific metadata handling for the JSON and TSV sidecar files
            typ = ft_senstype(varargin{1});
        end

    otherwise
        % assume that the file on disk contains raw timeseries data that can be read by FieldTrip
        if isequal(cfg.datatype, 'meg')
            need_meg_json = true;
        elseif isequal(cfg.datatype, 'eeg')
            need_eeg_json = true;
        elseif isequal(cfg.datatype, 'ieeg')
            need_ieeg_json = true;
        elseif isequal(cfg.datatype, 'emg')
            need_emg_json = true;
        elseif isequal(cfg.datatype, 'exg')
            need_exg_json = true;
        elseif isequal(cfg.datatype, 'nirs')
            need_nirs_json = true;
        elseif isequal(cfg.datatype, 'physio')
            need_physio_json = true;
        elseif isequal(cfg.datatype, 'stim')
            need_stim_json = true;
        elseif isequal(cfg.datatype, 'eyetracker')
            need_eyetracker_json = true;
        elseif isequal(cfg.datatype, 'motion')
            need_motion_json = true;
        elseif isequal(cfg.datatype, 'events')
            need_events_tsv = true;
        else
            ft_error('cannot determine the type of the data, please specify cfg.datatype');
        end

        % construct the low-level options as key-value pairs, these are passed to FT_READ_HEADER, FT_READ_DATA and FT_READ_EVENT
        headeropt = {};
        headeropt = ft_setopt(headeropt, 'headerformat',   ft_getopt(cfg, 'headerformat'));        % is passed to low-level function, empty implies autodetection
        headeropt = ft_setopt(headeropt, 'chantype',       ft_getopt(cfg, 'chantype', {}));        % 2017.10.10 AB required for NeuroOmega files
        headeropt = ft_setopt(headeropt, 'checkmaxfilter', false);
        headeropt = ft_setopt(headeropt, 'readbids',       false);

        dataopt   = {};
        dataopt   = ft_setopt(dataopt, 'checkboundary',    false);
        dataopt   = ft_setopt(dataopt, 'dataformat',       ft_getopt(cfg, 'dataformat'));          % is passed to low-level function, empty implies autodetection

        eventopt  = {};
        eventopt  = ft_setopt(eventopt, 'readbids',        false);
        eventopt  = ft_setopt(eventopt, 'eventformat',     ft_getopt(cfg, 'eventformat'));         % is passed to low-level function, empty implies autodetection

        if ~isempty(cfg.dataset)
            hdr = ft_read_header(cfg.headerfile, headeropt{:});
            if strcmp(cfg.method, 'convert')
                % the data should be converted and written to disk
                dat     = ft_read_data(cfg.datafile, 'header', hdr, 'begsample', 1, 'endsample', hdr.nSamples*hdr.nTrials, dataopt{:});
                trigger = ft_read_event(cfg.datafile, 'header', hdr, eventopt{:});
            end
        end

end % switch typ

if need_meg_json || need_eeg_json || need_ieeg_json
    % determine whether an electrode definition is available
    if isfield(cfg, 'elec') && ~isempty(cfg.elec)
        need_electrodes_tsv = true;
    elseif exist('hdr', 'var') && isfield(hdr, 'elec')
        need_electrodes_tsv = true;
    elseif ~isempty(varargin) && isfield(varargin{1}, 'elec') && ~isempty(varargin{1}.elec)
        need_electrodes_tsv = true;
    else
        need_electrodes_tsv = ~isequaln(cfg.electrodes.name, nan);
    end
end

if need_nirs_json
    % determine whether an optode definition is available
    if isfield(cfg, 'opto') && ~isempty(cfg.opto)
        need_optodes_tsv = true;
    elseif exist('hdr', 'var') && isfield(hdr, 'opto')
        need_optodes_tsv = true;
    elseif ~isempty(varargin) && isfield(varargin{1}, 'opto') && ~isempty(varargin{1}.opto)
        need_optodes_tsv = true;
    else
        need_optodes_tsv = ~isequaln(cfg.optodes.name, nan);
    end
end

need_events_tsv       = need_events_tsv       || need_meg_json || need_eeg_json || need_ieeg_json || need_emg_json || need_exg_json || need_nirs_json || need_eyetracker_json || need_motion_json || (contains(cfg.outputfile, 'task') || ~isempty(cfg.TaskName) || ~isempty(cfg.task)) || ~isempty(cfg.events);
need_channels_tsv     = need_channels_tsv     || need_meg_json || need_eeg_json || need_ieeg_json || need_emg_json || need_exg_json || need_nirs_json || need_motion_json ;
need_coordsystem_json = need_coordsystem_json || need_meg_json || need_electrodes_tsv || need_nirs_json ;

if need_emg_json
    ft_warning('EMG data is not yet part of the official BIDS specification');
    cfg.dataset_description.BIDSVersion = 'n/a';
elseif need_exg_json
    ft_warning('EXG data is not yet part of the official BIDS specification');
    cfg.dataset_description.BIDSVersion = 'n/a';
elseif need_audio_json
    ft_warning('audio data is not yet part of the official BIDS specification');
    cfg.dataset_description.BIDSVersion = 'n/a';
elseif need_video_json
    ft_warning('video data is not yet part of the official BIDS specification');
    cfg.dataset_description.BIDSVersion = 'n/a';
elseif need_eyetracker_json
    ft_warning('eyetracker data is not yet part of the official BIDS specification');
    cfg.dataset_description.BIDSVersion = 'n/a';
elseif need_motion_json
    ft_warning('motion data is not yet part of the official BIDS specification');
    cfg.dataset_description.BIDSVersion = 'n/a';
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% get the defaults and user-specified settings for each possible sidecar file
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.dataset_description);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
dataset_description_settings = keepfields(cfg.dataset_description, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
generic_settings = keepfields(cfg, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.mri);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
mri_settings = keepfields(cfg.mri, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.meg);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
meg_settings = keepfields(cfg.meg, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.eeg);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
eeg_settings = keepfields(cfg.eeg, fn);

% make the relevant selection, most json fields start with a capital letter, but some start with iEEG
fn = fieldnames(cfg.ieeg);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*|^iEEG')));
ieeg_settings = keepfields(cfg.ieeg, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.emg);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
emg_settings = keepfields(cfg.emg, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.exg);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
exg_settings = keepfields(cfg.exg, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.nirs);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
nirs_settings = keepfields(cfg.nirs, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.audio);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
audio_settings = keepfields(cfg.audio, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.video);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
video_settings = keepfields(cfg.video, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.physio);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
physio_settings = keepfields(cfg.physio, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.stim);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
stim_settings = keepfields(cfg.stim, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.eyetracker);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
eyetracker_settings = keepfields(cfg.eyetracker, fn);

% make the relevant selection, all json fields start with a capital letter
fn = fieldnames(cfg.motion);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*')));
motion_settings = keepfields(cfg.motion, fn);

% make the relevant selection, most json fields start with a capital letter, but some start with iEEG
fn = fieldnames(cfg.coordsystem);
fn = fn(~cellfun(@isempty, regexp(fn, '^[A-Z].*|^iEEG')));
coordsystem_settings = keepfields(cfg.coordsystem, fn);

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% construct the content for the json and tsv files
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

%% need_mri_json
if need_mri_json
    % start with the information from the DICOM header
    mri_json = keepfields(dcm, [fieldnames(mri_settings); fieldnames(generic_settings)]);
    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    mri_json = mergestruct(mri_settings,     mri_json, false);
    mri_json = mergestruct(generic_settings, mri_json, false);
end % if need_mri_json

%% need_meg_json
if need_meg_json
    meg_json.SamplingFrequency          = hdr.Fs;
    meg_json.MEGChannelCount            = sum(strcmpi(hdr.chantype, 'megmag') | strcmpi(hdr.chantype, 'meggrad') | strcmpi(hdr.chantype, 'megplanar') | strcmpi(hdr.chantype, 'megaxial'));
    meg_json.MEGREFChannelCount         = sum(strcmpi(hdr.chantype, 'refmag') | strcmpi(hdr.chantype, 'refgrad') | strcmpi(hdr.chantype, 'refplanar') | strcmpi(hdr.chantype, 'ref'));
    meg_json.EEGChannelCount            = sum(strcmpi(hdr.chantype, 'eeg'));
    meg_json.ECOGChannelCount           = sum(strcmpi(hdr.chantype, 'ecog'));
    meg_json.SEEGChannelCount           = sum(strcmpi(hdr.chantype, 'seeg') | strcmpi(hdr.chantype, 'dbs'));
    meg_json.EOGChannelCount            = sum(strcmpi(hdr.chantype, 'eog'));
    meg_json.ECGChannelCount            = sum(strcmpi(hdr.chantype, 'ecg'));
    meg_json.EMGChannelCount            = sum(strcmpi(hdr.chantype, 'emg'));
    meg_json.MiscChannelCount           = sum(strcmpi(hdr.chantype, 'misc') | strcmpi(hdr.chantype, 'unknown'));
    meg_json.TriggerChannelCount        = sum(contains(lower(hdr.chantype), 'trigger'));
    meg_json.RecordingDuration          = (hdr.nTrials*hdr.nSamples)/hdr.Fs;

    if hdr.nTrials>1
        meg_json.EpochLength              = hdr.nSamples/hdr.Fs;
    end
    if ft_senstype(hdr.grad, 'ctf151')
        meg_json.ContinuousHeadLocalization = any(strcmpi(hdr.chantype, 'headloc')); % CTF specific
        meg_json.Manufacturer             = 'CTF';
        meg_json.ManufacturersModelName   = 'CTF-151';
    elseif ft_senstype(hdr.grad, 'ctf275')
        meg_json.ContinuousHeadLocalization = any(strcmpi(hdr.chantype, 'headloc')); % CTF specific
        meg_json.Manufacturer             = 'CTF';
        meg_json.ManufacturersModelName   = 'CTF-275';
    elseif ft_senstype(hdr.grad, 'neuromag122')
        meg_json.Manufacturer             = 'Neuromag/Elekta/Megin';
        meg_json.ManufacturersModelName   = 'Neuromag-122';
    elseif ft_senstype(hdr.grad, 'neuromag306')
        meg_json.Manufacturer             = 'Neuromag/Elekta/Megin';
        % the ManufacturersModelName could be either Vectorview or Triux
    end

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    meg_json = mergestruct(meg_settings,     meg_json, false);
    meg_json = mergestruct(generic_settings, meg_json, false);
end % if need_meg_json

%% need_eeg_json
if need_eeg_json
    eeg_json.SamplingFrequency          = hdr.Fs;
    eeg_json.EEGChannelCount            = sum(strcmpi(hdr.chantype, 'eeg'));
    eeg_json.EOGChannelCount            = sum(strcmpi(hdr.chantype, 'eog'));
    eeg_json.ECGChannelCount            = sum(strcmpi(hdr.chantype, 'ecg'));
    eeg_json.EMGChannelCount            = sum(strcmpi(hdr.chantype, 'emg'));
    eeg_json.TriggerChannelCount        = sum(strcmpi(hdr.chantype, 'trigger'));
    eeg_json.MiscChannelCount           = sum(strcmpi(hdr.chantype, 'misc') | strcmpi(hdr.chantype, 'unknown'));
    eeg_json.RecordingDuration          = (hdr.nTrials*hdr.nSamples)/hdr.Fs;
    if hdr.nTrials>1
        eeg_json.EpochLength              = hdr.nSamples/hdr.Fs;
    end

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    eeg_json = mergestruct(eeg_settings,     eeg_json, false);
    eeg_json = mergestruct(generic_settings, eeg_json, false);
end % if need_eeg_json

%% need_ieeg_json
if need_ieeg_json
    ieeg_json.SamplingFrequency          = hdr.Fs;
    ieeg_json.ECOGChannelCount           = sum(strcmpi(hdr.chantype, 'ecog'));
    ieeg_json.SEEGChannelCount           = sum(strcmpi(hdr.chantype, 'seeg') | strcmpi(hdr.chantype, 'dbs'));
    ieeg_json.EEGChannelCount            = sum(strcmpi(hdr.chantype, 'eeg'));
    ieeg_json.EOGChannelCount            = sum(strcmpi(hdr.chantype, 'eog'));
    ieeg_json.ECGChannelCount            = sum(strcmpi(hdr.chantype, 'ecg'));
    ieeg_json.EMGChannelCount            = sum(strcmpi(hdr.chantype, 'emg'));
    ieeg_json.TriggerChannelCount        = sum(strcmpi(hdr.chantype, 'trigger'));
    ieeg_json.MiscChannelCount           = sum(strcmpi(hdr.chantype, 'misc') | strcmpi(hdr.chantype, 'unknown'));
    ieeg_json.RecordingDuration          = (hdr.nTrials*hdr.nSamples)/hdr.Fs;
    if hdr.nTrials>1
        ieeg_json.EpochLength              = hdr.nSamples/hdr.Fs;
    end

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    ieeg_json = mergestruct(ieeg_settings,    ieeg_json, false);
    ieeg_json = mergestruct(generic_settings, ieeg_json, false);
end

%% need_emg_json
if need_emg_json
    emg_json.SamplingFrequency          = hdr.Fs;
    emg_json.EOGChannelCount            = sum(strcmpi(hdr.chantype, 'eog'));
    emg_json.ECGChannelCount            = sum(strcmpi(hdr.chantype, 'ecg'));
    emg_json.EMGChannelCount            = sum(strcmpi(hdr.chantype, 'emg'));
    emg_json.TriggerChannelCount        = sum(strcmpi(hdr.chantype, 'trigger'));
    emg_json.MiscChannelCount           = sum(strcmpi(hdr.chantype, 'misc') | strcmpi(hdr.chantype, 'unknown'));
    emg_json.RecordingDuration          = (hdr.nTrials*hdr.nSamples)/hdr.Fs;
    if hdr.nTrials>1
        emg_json.EpochLength              = hdr.nSamples/hdr.Fs;
    end

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    emg_json = mergestruct(emg_settings,     emg_json, false);
    emg_json = mergestruct(generic_settings, emg_json, false);
end

%% need_exg_json
if need_exg_json
    exg_json.SamplingFrequency          = hdr.Fs;
    exg_json.RecordingDuration          = (hdr.nTrials*hdr.nSamples)/hdr.Fs;
    if hdr.nTrials>1
        exg_json.EpochLength              = hdr.nSamples/hdr.Fs;
    end

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    exg_json = mergestruct(exg_settings,     exg_json, false);
    exg_json = mergestruct(generic_settings, exg_json, false);
end

%% need_nirs_json
if need_nirs_json
    nirs_json.SamplingFrequency         = hdr.Fs;
    nirs_json.RecordingDuration         = (hdr.nTrials*hdr.nSamples)/hdr.Fs;
    if hdr.nTrials>1
        nirs_json.EpochLength             = hdr.nSamples/hdr.Fs;
    end
    nirs_json.NIRSChannelCount          = sum(strcmpi(hdr.chantype, 'nirs'));
    nirs_json.ACCELChannelCount         = sum(strcmpi(hdr.chantype, 'accel'));
    nirs_json.GYROChannelCount          = sum(strcmpi(hdr.chantype, 'gyro'));
    nirs_json.MAGNChannelCount          = sum(strcmpi(hdr.chantype, 'magn'));
    nirs_json.MISCChannelCount          = sum(strcmpi(hdr.chantype, 'misc') | strcmpi(hdr.chantype, 'unknown') | strcmpi(hdr.chantype, 'aux'));
    [opto_labels, opto_idx]             = unique(hdr.opto.optolabel); % select unique optodes
    nirs_json.NIRSSourceOptodeCount     = sum(strcmpi(hdr.opto.optotype(opto_idx), 'transmitter'));
    nirs_json.NIRSDetectorOptodeCount   = sum (strcmpi(hdr.opto.optotype(opto_idx), 'receiver'));

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    nirs_json = mergestruct(nirs_settings,    nirs_json, false);
    nirs_json = mergestruct(generic_settings, nirs_json, false);
end

%% need_audio_json
if need_audio_json
    audio_json.SampleRate         = audio.SampleRate;
    audio_json.RecordingDuration  = audio.Duration; % please note that this is not consistent with VideoDuration/AudioDuration for a video file, but it is consistent with MEG/EEG/iEEG
    audio_json.ChannelCount       = audio.NumChannels;

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    audio_json = mergestruct(audio_settings,   audio_json, false);
    audio_json = mergestruct(generic_settings, audio_json, false);
end

%% need_video_json
if need_video_json
    video_json.FrameRate          = video.FrameRate;
    video_json.Width              = video.Width;
    video_json.Height             = video.Height;
    video_json.VideoDuration      = video.Duration; % to distinguish it from the audio duration
    video_json.AudioDuration      = audio.Duration; % to distinguish it from the video duration
    video_json.AudioSampleRate    = audio.SampleRate;
    video_json.AudioChannelCount  = audio.NumChannels;

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    video_json = mergestruct(video_settings,   video_json, false);
    video_json = mergestruct(generic_settings, video_json, false);
end

%% need_physio_json
if need_physio_json
    physio_json.SamplingFrequency = hdr.Fs;
    physio_json.StartTime = nan;
    physio_json.Columns = hdr.label;

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    physio_json = mergestruct(physio_settings,  physio_json, false);
    physio_json = mergestruct(generic_settings, physio_json, false);
end

%% need_stim_json
if need_stim_json
    stim_json.SamplingFrequency = hdr.Fs;
    stim_json.StartTime = nan;
    stim_json.Columns = hdr.label;

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    stim_json = mergestruct(stim_settings,    stim_json, false);
    stim_json = mergestruct(generic_settings, stim_json, false);
end

%% need_eyetracker_json
if need_eyetracker_json
    eyetracker_json.SamplingFrequency = hdr.Fs;
    eyetracker_json.StartTime = nan;
    eyetracker_json.Columns = hdr.label;

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    eyetracker_json = mergestruct(eyetracker_settings,  eyetracker_json, false);
    eyetracker_json = mergestruct(generic_settings,     eyetracker_json, false);
end

%% need_motion_json
if need_motion_json
    motion_json.SamplingFrequency     = hdr.Fs;
    motion_json.StartTime             = nan;
    motion_json.MotionChannelCount    = hdr.nChans;
    motion_json.RecordingDuration     = (hdr.nSamples*hdr.nTrials)/hdr.Fs;
    motion_json.SamplingFrequencyEffective = size(dat,2)/motion_json.RecordingDuration;
    motion_json.POSChannelCount       = sum(strcmpi(hdr.chantype, 'POS'));
    motion_json.ORNTChannelCount      = sum(strcmpi(hdr.chantype, 'ORNT'));
    motion_json.VELChannelCount       = sum(strcmpi(hdr.chantype, 'VEL'));
    motion_json.ANGVELChannelCount    = sum(strcmpi(hdr.chantype, 'ANGVEL'));
    motion_json.ACCChannelCount       = sum(strcmpi(hdr.chantype, 'ACC'));
    motion_json.ANGACCChannelCount    = sum(strcmpi(hdr.chantype, 'ANGACC'));
    motion_json.MAGNChannelCount      = sum(strcmpi(hdr.chantype, 'MAGN'));
    motion_json.JNTANGChannelCount    = sum(strcmpi(hdr.chantype, 'JNTANG'));
    if isfield(cfg, 'channels') && isfield(cfg.channels, 'tracked_point')
        motion_json.TrackedPointsCount  = numel(setdiff(unique(cfg.channels.tracked_point), 'n/a'));
    end

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    motion_json = mergestruct(motion_settings,  motion_json, false);
    motion_json = mergestruct(generic_settings, motion_json, false);

end % if need_motion_json

%% need_channels_tsv
if need_channels_tsv

    if isstruct(cfg.channels)
        % remove fields with non-informative defaults
        fn = fieldnames(cfg.channels);
        for i=1:numel(fn)
            if isequaln(cfg.channels.(fn{i}), nan)
                % a single nan means that it was set as default
                cfg.channels = rmfield(cfg.channels, fn{i});
            end
        end
        try
            cfg.channels = convert_table(cfg.channels);
        catch
            ft_error('incorrect specification of cfg.channels');
        end
    end

    % channel information can come from the header and from cfg.channels
    channels_tsv = hdr2table(hdr);
    channels_tsv = mergetable(channels_tsv, cfg.channels, 'name');

    % columns should appear in a specific order
    if need_nirs_json
        required = {'name', 'type', 'source', 'detector', 'wavelength_nominal', 'units'};
    else
        required = {'name', 'type', 'units', 'low_cutoff', 'high_cutoff'};
    end
    optional = setdiff(channels_tsv.Properties.VariableNames, required, 'stable');
    channels_tsv = sort_columns(channels_tsv, [required, optional]);

    % the default for cfg.channels consists of one row where all values are nan, this needs to be removed
    keep = false(size(channels_tsv.name));
    for i=1:numel(channels_tsv.name)
        keep(i) = ischar(channels_tsv.name{i});
    end
    channels_tsv = channels_tsv(keep,:);

    % there are some chanel types used in FieldTrip that are named differently in BIDS
    channels_tsv.type(strcmpi(channels_tsv.type, 'unknown'))     = {'OTHER'};
    channels_tsv.type(strcmpi(channels_tsv.type, 'clock'))       = {'SYSCLOCK'};
    channels_tsv.type(strcmpi(channels_tsv.type, 'meggrad'))     = {'MEGGRADAXIAL'};
    channels_tsv.type(strcmpi(channels_tsv.type, 'megplanar'))   = {'MEGGRADPLANAR'};
    channels_tsv.type(strcmpi(channels_tsv.type, 'refmag'))      = {'MEGREFMAG'};
    channels_tsv.type(strcmpi(channels_tsv.type, 'refgrad'))     = {'MEGREFGRADAXIAL'};
    channels_tsv.type(strcmpi(channels_tsv.type, 'refplanar'))   = {'MEGREFGRADPLANAR'};
    channels_tsv.type(strcmpi(channels_tsv.type, 'respiration')) = {'RESP'};
    channels_tsv.type(strcmpi(channels_tsv.type, 'headloc'))     = {'HLU'};
    channels_tsv.type(strcmpi(channels_tsv.type, 'headloc_gof')) = {'FITERR'};
    channels_tsv.type(strcmpi(channels_tsv.type, 'ori'))         = {'ORNT'};
    % trigger, analog trigger, and digital trigger all have to be renamed to TRIG
    channels_tsv.type(contains(channels_tsv.type, 'trigger', 'IgnoreCase', true)) = {'TRIG'};
    channels_tsv.type(contains(channels_tsv.type, 'nirs'))       = {'unknown'}; % depends on the type of measurement and should be provided by the user
    % channel types in BIDS must be in upper case
    channels_tsv.type = upper(channels_tsv.type);

    % do a sanity check on the number of channels for the electrophysiology data types
    if need_meg_json
        type_json = meg_json;
    elseif need_eeg_json
        type_json = eeg_json;
    elseif need_ieeg_json
        type_json = ieeg_json;
    elseif need_emg_json
        type_json = emg_json;
    elseif need_exg_json
        type_json = exg_json;
    elseif need_nirs_json
        type_json = nirs_json;
    elseif need_motion_json
        type_json = motion_json;
    end
    fn = fieldnames(type_json);
    fn = fn(endsWith(fn, 'ChannelCount') & ~contains(fn,'ShortChannel'));
    jsoncount = 0;
    for i=1:numel(fn)
        if ~isempty(type_json.(fn{i}))
            jsoncount = jsoncount + type_json.(fn{i});
        end
    end
    if size(channels_tsv,1)~=jsoncount
        ft_warning('inconsistent specification of the channel count: %d in the json, %d in the tsv', jsoncount, size(channels_tsv,1));
    end
end % if need_channels_tsv

%% need_electrodes_tsv
if need_electrodes_tsv

    % try to get the elec structure from the configuration or data
    try
        tmpcfg = keepfields(cfg, {'elec'});
        tmpcfg.senstype = 'eeg';
        if ~isempty(varargin)
            elec = ft_fetch_sens(tmpcfg, varargin{1});
        elseif exist('hdr', 'var') && isfield(hdr, 'elec')
            elec = hdr.elec;
        else
            elec = ft_fetch_sens(tmpcfg);
        end
    catch
        elec = [];
    end

    if isstruct(cfg.electrodes)
        % remove fields with non-informative defaults
        fn = fieldnames(cfg.electrodes);
        for i=1:numel(fn)
            if isequaln(cfg.electrodes.(fn{i}), nan)
                % a single nan means that it was set as default
                cfg.electrodes = rmfield(cfg.electrodes, fn{i});
            end
        end
        try
            cfg.electrodes = convert_table(cfg.electrodes);
        catch
            ft_error('incorrect specification of cfg.electrodes.%s', fn{i});
        end
    end

    % electrode details can be specified in cfg.elec, data.elec or in cfg.electrodes
    electrodes_tsv = elec2table(elec);                                    % this includes the cfg.elec and data.elec
    electrodes_tsv = mergetable(electrodes_tsv, cfg.electrodes, 'name'); % this includes the cfg.electrodes

    % the default for cfg.electrodes consists of one row where all values are nan, this needs to be removed
    keep = false(size(electrodes_tsv.name));
    for i=1:numel(electrodes_tsv.name)
        keep(i) = ischar(electrodes_tsv.name{i});
    end
    electrodes_tsv = electrodes_tsv(keep,:);

end % need_electrodes_tsv

%% need_optodes_tsv
if need_optodes_tsv

    % try to get the opto structure from the configuration or data
    try
        tmpcfg = keepfields(cfg, {'opto'});
        tmpcfg.senstype='nirs';
        if ~isempty(varargin)
            opto = ft_fetch_sens(tmpcfg, varargin{1});
        elseif exist('hdr', 'var') && isfield(hdr, 'opto')
            opto = hdr.opto;
        else
            opto = ft_fetch_sens(tmpcfg);
        end
    catch
        opto = [];
    end

    if isstruct(cfg.optodes)
        % remove fields with non-informative defaults
        fn = fieldnames(cfg.optodes);
        for i=1:numel(fn)
            if isequaln(cfg.optodes.(fn{i}), nan)
                % a single nan means that it was set as default
                cfg.optodes = rmfield(cfg.optodes, fn{i});
            end
        end
        try
            cfg.optodes = convert_table(cfg.optodes);
        catch
            ft_error('incorrect specification of cfg.optodes.');
        end
    end

    % optode details can be specified in cfg.opto, data.opto or cfg.optodes
    optodes_tsv = opto2table(opto);                              % this includes the cfg.opto and data.opto
    optodes_tsv = mergetable(optodes_tsv, cfg.optodes, 'name'); % this includes the cfg.optodes

    % the default for cfg.electrodes consists of one row where all values are nan, this needs to be removed
    keep = false(size(optodes_tsv.name));
    for i=1:numel(optodes_tsv.name)
        keep(i) = ischar(optodes_tsv.name{i});
    end
    optodes_tsv = optodes_tsv(keep,:);

end % need_optodes_tsv

%% need_coordsystem_json
if need_coordsystem_json
    if isfield(hdr, 'grad') && ft_senstype(hdr.grad, 'ctf')

        % CTF empty-room recordings use the standard positions of the coils, as if they were measured
        emptyroom = all(hdr.orig.hc.standard == hdr.orig.hc.dewar, [1 2]);
        if emptyroom
            ft_notice('this seems to be an empty room recording')
        end

        if ~emptyroom
            % coordinate system for MEG sensors
            coordsystem_json.MEGCoordinateSystem                 = 'CTF';
            coordsystem_json.MEGCoordinateUnits                  = 'cm';
            coordsystem_json.MEGCoordinateSystemDescription      = 'CTF head coordinates, orientation ALS, origin between the ears';
            % coordinate system for head localization coils
            coordsystem_json.HeadCoilCoordinates                 = []; % see below
            coordsystem_json.HeadCoilCoordinateSystem            = 'CTF';
            coordsystem_json.HeadCoilCoordinateUnits             = 'cm';
            coordsystem_json.HeadCoilCoordinateSystemDescription = 'CTF head coordinates, orientation ALS, origin between the ears';
            if isempty(coordsystem_json.HeadCoilCoordinates)
                % get the positions from the dataset header
                label = cellstr(hdr.orig.hc.names);
                position = hdr.orig.hc.head;
                for i=1:numel(label)
                    coordsystem_json.HeadCoilCoordinates.(fixname(label{i})) = position(:,i)';
                end
            end
        else
            % coordinate system for MEG sensors
            coordsystem_json.MEGCoordinateSystem                 = 'Other';
            coordsystem_json.MEGCoordinateUnits                  = 'cm';
            coordsystem_json.MEGCoordinateSystemDescription      = 'CTF head coordinates with standard coil positions relative to dewar, orientation ALS';
        end % if ~emptyroom

    elseif isfield(hdr, 'grad') && ft_senstype(hdr.grad, 'neuromag')

        % Neuromag empty-room recordings do not have digitizer information
        emptyroom = isempty(hdr.orig.dig);
        if emptyroom
            ft_notice('this seems to be an empty room recording')
        end

        if ~emptyroom
            % coordinate system for MEG sensors
            coordsystem_json.MEGCoordinateSystem                 = 'ElektaNeuromag';
            coordsystem_json.MEGCoordinateUnits                  = 'm';
            coordsystem_json.MEGCoordinateSystemDescription      = 'ElektaNeuromag head coordinates, orientation RAS, origin between the ears';
            % coordinate system for head localization coils
            coordsystem_json.HeadCoilCoordinates                 = []; % see below
            coordsystem_json.HeadCoilCoordinateSystem            = 'ElektaNeuromag';
            coordsystem_json.HeadCoilCoordinateUnits             = 'm';
            coordsystem_json.HeadCoilCoordinateSystemDescription = 'ElektaNeuromag head coordinates, orientation RAS, origin between the ears';
            if isempty(coordsystem_json.HeadCoilCoordinates)
                idxHPI = find([hdr.orig.dig.kind] == 2); % count the kind==2 (HLU in the Elekta/Megin system), usually 4 or 5
                for i=1:length(idxHPI)
                    % the coils do not have a name but are numbered
                    coordsystem_json.HeadCoilCoordinates.(['coil' num2str(i)]) = hdr.orig.dig(idxHPI(i)).r';
                end
            end
            % coordinates of the anatomical landmarks (LPA/RPA/NAS)
            coordsystem_json.AnatomicalLandmarkCoordinates                 = []; % see below
            coordsystem_json.AnatomicalLandmarkCoordinateSystem            = 'ElektaNeuromag';
            coordsystem_json.AnatomicalLandmarkCoordinateUnits             = 'm';
            coordsystem_json.AnatomicalLandmarkCoordinateSystemDescription = 'ElektaNeuromag head coordinates, orientation RAS, origin between the ears';
            if isempty(coordsystem_json.AnatomicalLandmarkCoordinates)
                coordsystem_json = rmfield(coordsystem_json, 'AnatomicalLandmarkCoordinates'); % needed to set the names afterwards
                coordsystem_json.AnatomicalLandmarkCoordinates.lpa = hdr.orig.dig(1).r';
                coordsystem_json.AnatomicalLandmarkCoordinates.rpa = hdr.orig.dig(2).r';
                coordsystem_json.AnatomicalLandmarkCoordinates.nas = hdr.orig.dig(3).r';
            else
                coordsystem_json.AnatomicalLandmarkCoordinates = nan;
            end
        else
            % coordinate system for MEG sensors
            coordsystem_json.MEGCoordinateSystem                 = 'Other';
            coordsystem_json.MEGCoordinateUnits                  = 'm';
            coordsystem_json.MEGCoordinateSystemDescription      = 'ElektaNeuromag dewar coordinates, orientation RAS';
        end % if ~emptyroom

    else
        ft_warning('automatic coordsystem handling not yet supported for this data, you MUST specify cfg.coordsystem');
        coordsystem_json = table();
    end

    % merge the information specified by the user with that from the data
    % in case fields appear in both, the first input overrules the second
    coordsystem_json = mergestruct(coordsystem_settings, coordsystem_json, false); % FIXME the order of precedence is different here
end % if need_coordsystem_json



%% need_events_tsv
if need_events_tsv

    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    %% align the presentation events with BOLD volumes or MEG/EEG triggers
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

    if ~isempty(cfg.presentationfile) && need_mri_json
        % the code can either align presentation and BOLD volumes, or use the user-supplied events that must be aligned
        assert(isempty(cfg.events) || (isstruct(cfg.events) && numel(fieldnames(cfg.events))==0), 'this is mutually exclusive with specifying cfg.events')

        % read the events from the presentation file
        presentation = ft_read_event(cfg.presentationfile);

        % merge the information with the json sidecar file
        % in case fields appear in both, the first input overrules the second
        tmp = mergestruct(mri_json, ft_read_json(corresponding_json(cfg.outputfile)), false);
        assert(~isempty(tmp.RepetitionTime), 'you must specify cfg.mri.RepetitionTime');

        % create a header structure that represents the fMRI timeseries
        hdr.Fs = 1/tmp.RepetitionTime;
        hdr.nSamples = size(mri.anatomy, 4);

        % create a event structure with one trigger for each BOLD volume
        trigger = [];
        for i=1:hdr.nSamples
            trigger(i).type   = 'volume';
            trigger(i).sample = i;
        end

        % align the presentation events with the triggers
        cfg.events = align_presentation(presentation, cfg.presentation, trigger, cfg.trigger, hdr, istrue(cfg.feedback));

    elseif ~isempty(cfg.presentationfile) && (need_meg_json || need_eeg_json || need_ieeg_json || need_emg_json || need_exg_json || need_nirs_json)
        % the code can either align presentation and trigger channel, or use the user-supplied events that must be aligned
        assert(isempty(cfg.events) || (isstruct(cfg.events) && numel(fieldnames(cfg.events))==0), 'this is mutually exclusive with specifying cfg.events')

        % read the events from the presentation file
        presentation = ft_read_event(cfg.presentationfile);

        % align the presentation events with the triggers
        cfg.events = align_presentation(presentation, cfg.presentation, trigger, cfg.trigger, hdr, istrue(cfg.feedback));

        % also include all triggers from the MEG/EEG dataset as events
        cfg.events = appendstruct(cfg.events, trigger);

    end % if presentationfile

    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    %% convert the events into a table
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

    if istable(cfg.events) && all(ismember({'onset', 'duration'}, fieldnames(cfg.events)))
        % use the events table as it is
        events_tsv = cfg.events;

    elseif istable(cfg.events) && all(ismember({'begsample', 'endsample', 'offset'}, fieldnames(cfg.events)))
        % it is a "trl" matrix formatted as table, use it as it is, but add onset and duration
        events_tsv = cfg.events;
        begsample                   = table2array(events_tsv(:,{'begsample'}));
        endsample                   = table2array(events_tsv(:,{'endsample'}));
        onset                       = (begsample-1)./hdr.Fs;
        duration                    = (endsample-begsample+1)./hdr.Fs;
        table_onset_duration        = table(onset, duration);
        events_tsv                  = [table_onset_duration events_tsv];

    elseif istable(cfg.events) && ~isempty(cfg.events)
        ft_error('cannot interpret cfg.events');

    elseif isstruct(cfg.events) && ~isempty(cfg.events) && numel(fieldnames(cfg.events))>0
        % it is the structure output from FT_READ_EVENT
        if exist('hdr', 'var')
            events_tsv = event2table(hdr, cfg.events);
        else
            events_tsv = event2table([], cfg.events);
        end

    elseif isnumeric(cfg.events) && ~isempty(cfg.events)
        % it is a "trl" matrix formatted as numeric array, convert it to an events table
        begsample = cfg.events(:,1);
        endsample = cfg.events(:,2);
        offset    = cfg.events(:,3); % this is not used for the events.tsv
        if size(cfg.events, 2)>3
            ft_warning('additional columns in the trl matrix are ignored');
        end
        % convert to the required fields
        onset     = (begsample-1)/hdr.Fs;
        duration  = (endsample-begsample+1)/hdr.Fs;
        events_tsv = table(onset, duration, begsample, endsample, offset);

    elseif isempty(cfg.events) && exist('trigger', 'var')
        % convert the triggers from FT_READ_EVENT into a table
        if exist('hdr', 'var')
            events_tsv = event2table(hdr, trigger);
        else
            events_tsv = event2table([], trigger);
        end

    elseif ~isempty(cfg.presentationfile)
        % read the presentation file and convert into a table
        events_tsv = event2table([], ft_read_event(cfg.presentationfile));

    else
        ft_warning('no events were specified');
        % make an empty table with columns for onset and duration
        onset                   = [];
        duration                = [];
        begsample               = [];
        endsample               = [];
        offset                  = [];

        events_tsv = table(onset, duration, begsample, endsample, offset);
    end

    if isempty(events_tsv)
        ft_warning('no events found');
        need_events_tsv = false;

    else
        % ensure that column names are in lower case
        events_tsv.Properties.VariableNames = lower(events_tsv.Properties.VariableNames);

        % ensure that the onset and duration appear as the first two columns
        required = {'onset', 'duration'};
        optional = setdiff(events_tsv.Properties.VariableNames, required, 'stable');
        events_tsv = sort_columns(events_tsv, [required, optional]);

        % sort the events ascending on the onset
        events_tsv = sortrows(events_tsv, 'onset');
    end

    if ~isempty(cfg.presentationfile) && need_mri_json
        % rename the column 'sample' into 'volume'
        events_tsv.Properties.VariableNames('sample') = {'volume'};
    end

end % if need_events_tsv

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% write or copy the data to the output file
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

[p, f, x] = fileparts(cfg.outputfile);
isdir_or_mkdir(p);

switch cfg.method
    case 'decorate'
        % look at the user's specification of cfg.datatype
        if strcmp(cfg.datatype, 'events')
            % add the TSV file extension, this is needed for behavioral data represented in scans.tsv
            % the events.tsv file will be written further down
            [p, f, x] = fileparts(cfg.outputfile);
            cfg.outputfile = fullfile(p, [f '.tsv']);
        end

    case 'convert'
        % the output depends on the type of input data
        switch typ
            case {'nifti', 'nifti2', 'nifti_fsl', 'dicom', 'volume'}
                [p, f, x] = fileparts(cfg.outputfile);
                if ~isequal(x, '.nii')
                    cfg.outputfile = fullfile(p, [f '.nii']);
                end
                if istrue(cfg.deface)
                    % deface the anatomical MRI
                    ft_info('defacing anatomical MRI');
                    if istrue(cfg.feedback)
                        tmpcfg = [];
                        tmpcfg.figurename = sprintf('ORIGINAL - %s', f);
                        tmpcfg.showcallinfo = 'no';
                        tmpcfg.colorbar = 'no';
                        ft_sourceplot(tmpcfg, mri);
                    end
                    tmpcfg = [];
                    tmpcfg.method = 'spm';
                    if ~isfield(mri, 'coordsys')
                        % try to determine it from the BIDS filename
                        coordsys = get_token(cfg.outputfile, 'space');
                        if ~isempty(coordsys)
                            mri.coordsys = coordsys;
                        else
                            mri.coordsys = 'unknown';
                        end
                    end
                    mri = ft_defacevolume(tmpcfg, mri);
                    if istrue(cfg.feedback)
                        tmpcfg = [];
                        tmpcfg.figurename = sprintf('DEFACED - %s', f);
                        tmpcfg.showcallinfo = 'no';
                        tmpcfg.colorbar = 'no';
                        ft_sourceplot(tmpcfg, mri);
                    end
                end
                ft_info('writing ''%s''\n', cfg.outputfile);
                ft_write_mri(cfg.outputfile, mri, 'dataformat', 'nifti');

            case {'ctf_ds', 'ctf_meg4', 'ctf_res4', 'ctf151', 'ctf275', 'neuromag_fif', 'neuromag122', 'neuromag306'}
                % we cannot write MEG data back to its native format
                ft_error('please use a system specific tool for converting MEG datasets');

            case {'presentation_log'}
                % the events.tsv file will be written further down
                [p, f, x] = fileparts(cfg.outputfile);
                cfg.outputfile = fullfile(p, [f '.tsv']);

            otherwise
                % look at the user's specification of cfg.datatype
                switch cfg.datatype
                    case {'eeg', 'ieeg', 'emg', 'exg'}
                        % write the data in BrainVision core file format
                        [p, f, x] = fileparts(cfg.outputfile);
                        cfg.outputfile = fullfile(p, [f '.vhdr']);

                        ft_info('writing ''%s''\n', cfg.outputfile);
                        ft_write_data(cfg.outputfile, dat, 'dataformat', 'brainvision_eeg', 'header', hdr, 'event', trigger);
                    case {'nirs'}
                        % write the data in SNIRF file format
                        [p, f, x] = fileparts(cfg.outputfile);
                        cfg.outputfile = fullfile(p, [f '.snirf']);
                        ft_info('writing ''%s''\n', cfg.outputfile);
                        ft_write_data(cfg.outputfile, dat, 'dataformat', 'snirf', 'header', hdr, 'event', trigger);
                    case {'physio', 'stim', 'eyetracker', 'motion'}
                        % write the data according to the Stim and Physio format as specified at
                        % https://bids-specification.readthedocs.io/en/stable/04-modality-specific-files/06-physiological-and-other-continuous-recordings.html
                        [p, f, x] = fileparts(cfg.outputfile);
                        cfg.outputfile = fullfile(p, [f '.tsv']);

                        ft_info('writing %s\n', cfg.outputfile);

                        if any(strcmp(cfg.datatype, {'motion', 'physio'}))
                            % with headers, the JSON will be written further down
                            writecell(hdr.label', cfg.outputfile, 'FileType', 'text', 'Delimiter', '\t');
                            writematrix(dat', cfg.outputfile, 'FileType', 'text', 'Delimiter', '\t', 'WriteMode','append');
                        else
                            % without headers, the JSON will be written further down
                            writematrix(dat', cfg.outputfile, 'FileType', 'text', 'Delimiter', '\t');
                        end

                    case {'events'}
                        % add the TSV file extension, this is needed for behavioral data represented in scans.tsv
                        % the events.tsv file will be written further down
                        [p, f, x] = fileparts(cfg.outputfile);
                        cfg.outputfile = fullfile(p, [f '.tsv']);
                    otherwise
                        ft_error('cannot determine how to write the data')
                end
        end % switch typ

    case 'copy'
        [p, f, xin] = fileparts(cfg.dataset);
        [p, f, xout] = fileparts(cfg.outputfile);
        if ~strcmp(xin, xout)
            ft_error('input and output filename extension do not match');
        end

        switch typ
            case {'dicom'}
                ft_error('DICOM files must be converted to NIfTI for BIDS compliance');

            case {'eyelink_edf', 'eyelink_asc'}
                ft_error('Eyelink files must be converted to TSV for BIDS compliance');

            case {'ctf_ds', 'ctf_meg4', 'ctf_res4', 'ctf151', 'ctf275'}
                % the data consists of a directory with multiple files inside
                ft_info('copying ''%s'' to ''%s''\n', cfg.dataset, cfg.outputfile);
                copy_ctf_files(cfg.dataset, cfg.outputfile, false);

            case {'brainvision_vhdr', 'brainvision_vmrk', 'brainvision_eeg', 'brainvision_dat', 'brainvision_seg'}
                % the data consists of three files and the header file contains pointers to the markers and data
                ft_info('copying ''%s'' to ''%s''\n', cfg.dataset, cfg.outputfile);
                copy_brainvision_files(cfg.dataset, cfg.outputfile, false);

            otherwise
                ft_info('copying ''%s'' to ''%s''\n', cfg.dataset, cfg.outputfile);
                copyfile(cfg.dataset, cfg.outputfile);
        end

    otherwise
        ft_error('unsupported value for cfg.method')
end % switch method


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% write the metadata to the json files
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% each of these has a corresponding json file
modality = {'mri', 'meg', 'eeg', 'ieeg', 'nirs', 'physio', 'stim', 'emg', 'exg', 'audio', 'video', 'eyetracker', 'motion', 'coordsystem'};
for i=1:numel(modality)
    if eval(sprintf('need_%s_json', modality{i}))
        modality_json = eval(sprintf('%s_json', modality{i}));
        modality_json = remove_empty(modality_json);

        if strcmp(modality{i}, 'coordsystem')
            [p, f] = fileparts(cfg.outputfile);
            f = remove_entity(f, 'task');     % remove _task-something
            f = remove_entity(f, 'acq');      % remove _acq-something
            f = remove_entity(f, 'ce');       % remove _ce-something
            f = remove_entity(f, 'rec');      % remove _rec-something
            f = remove_entity(f, 'dir');      % remove _dir-something
            f = remove_entity(f, 'run');      % remove _run-something
            f = remove_entity(f, 'mod');      % remove _mod-something
            f = remove_entity(f, 'echo');     % remove _echo-something
            f = remove_entity(f, 'proc');     % remove _proc-something
            f = remove_entity(f, 'desc');     % remove _desc-something
            f = remove_datatype(f);           % remove _meg, _eeg, etc.
            if ismember(modality{i}, {'mri', 'meg', 'motion', 'coordsystem'})
                f = add_entity(f, 'space', cfg.space);
            end
            filename = fullfile(p, [f '_coordsystem.json']);
        else
            % just replace the extension with json
            filename = corresponding_json(cfg.outputfile);
        end

        if isfile(filename)
            existing = ft_read_json(filename);
        else
            existing = [];
        end

        switch cfg.writejson
            case 'yes'
                if ~isempty(existing)
                    ft_warning('not overwriting the existing and non-empty file ''%s''', filename);
                else
                    ft_write_json(filename, modality_json);
                end
            case 'replace'
                ft_write_json(filename, modality_json);
            case 'merge'
                ft_write_json(filename, mergestruct(modality_json, existing, false))
            case 'no'
                % do nothing
                ft_info('not writing ''%s''\n', filename);
            otherwise
                ft_error('incorrect option for cfg.writejson');
        end % switch writejson
    end % if need_xxx_json
end % for each modality


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% write the metadata to the tsv files
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

% each of these has a corresponding tsv file
modality = {'channels', 'electrodes', 'optodes', 'events'};
for i=1:numel(modality)
    if eval(sprintf('need_%s_tsv', modality{i}))
        modality_tsv = eval(sprintf('%s_tsv', modality{i}));
        modality_tsv = remove_empty(modality_tsv);

        if any(strcmp(modality{i}, {'electrodes', 'optodes'}))
            [p, f] = fileparts(cfg.outputfile);
            f = remove_entity(f, 'task');     % remove _task-something
            f = remove_entity(f, 'acq');      % remove _acq-something
            f = remove_entity(f, 'tracksys'); % remove _tracksys-something
            f = remove_entity(f, 'ce');       % remove _ce-something
            f = remove_entity(f, 'rec');      % remove _rec-something
            f = remove_entity(f, 'dir');      % remove _dir-something
            f = remove_entity(f, 'run');      % remove _run-something
            f = remove_entity(f, 'mod');      % remove _mod-something
            f = remove_entity(f, 'echo');     % remove _echo-something
            f = remove_entity(f, 'proc');     % remove _proc-something
            f = remove_entity(f, 'desc');     % remove _desc-something
            f = remove_datatype(f);           % remove _meg, _eeg, etc.
            if ismember(modality{i}, {'electrodes', 'optodes'})
                f = add_entity(f, 'space', cfg.space);
            end
            filename = fullfile(p, sprintf('%s_%s.tsv', f, modality{i}));
        else
            [p, f] = fileparts(cfg.outputfile);
            f = remove_datatype(f); % remove _bold, _meg, etc.
            filename = fullfile(p, sprintf('%s_%s.tsv', f, modality{i}));
        end

        if isfile(filename)
            existing = ft_read_tsv(filename);
        else
            existing = [];
        end

        switch cfg.writetsv
            case 'yes'
                if ~isempty(existing)
                    ft_warning('not overwriting the existing and non-empty file ''%s''', filename);
                else
                    ft_write_tsv(filename, modality_tsv);
                end
            case 'replace'
                ft_write_tsv(filename, modality_tsv);
            case 'merge'
                if ~isempty(existing)
                    modality_tsv  = output_compatible(modality_tsv);
                    existing      = output_compatible(existing);
                    if strcmp(modality{i}, 'events')
                        % merge complete rows
                        modality_tsv = mergetable(modality_tsv, existing);
                    else
                        % use the channel name as the unique key
                        modality_tsv = mergetable(modality_tsv, existing, 'name');
                    end
                end
                ft_write_tsv(filename, modality_tsv);
            case 'no'
                % do nothing
                ft_info('not writing ''%s''\n', filename);
            otherwise
                ft_error('incorrect option for cfg.writetsv');
        end % switch
    end % if needd_xxx_tsv
end % for each modality


if ~isempty(cfg.bidsroot)
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    % update the dataset_description
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    filename = fullfile(cfg.bidsroot, 'dataset_description.json');

    if isfile(filename)
        existing = ft_read_json(filename);
    else
        existing = [];
    end

    switch cfg.writejson
        case 'yes'
            if ~isempty(existing)
                ft_warning('not overwriting the existing and non-empty file ''%s''', filename);
            else
                ft_write_json(filename, dataset_description_settings);
            end
        case 'replace'
            ft_write_json(filename, dataset_description_settings);
        case 'merge'
            ft_write_json(filename, mergestruct(dataset_description_settings, existing, false));
        case 'no'
            % do nothing
        otherwise
            ft_error('incorrect option for cfg.writejson');
    end % switch

    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    % update the participants.tsv
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

    this = table();
    this.participant_id = {['sub-' cfg.sub]};
    fn = fieldnames(cfg.participants);
    for i=1:numel(fn)
        % write [] as 'n/a'
        % write nan as 'n/a'
        % write boolean as 'True' or 'False'
        this.(fn{i}) = output_compatible(cfg.participants.(fn{i}));
    end

    filename = fullfile(cfg.bidsroot, 'participants.tsv');
    if isfile(filename)
        participants = ft_read_tsv(filename);
        % the participants.tsv is always merged
        participants = mergetable(participants, this, 'participant_id');
    else
        participants = this;
    end

    % write the updated file back to disk
    ft_write_tsv(filename, participants);

    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    % update the scans.tsv
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    if ~isempty(cfg.ses)
        % construct the output filename, with session directory
        filename = fullfile(cfg.bidsroot, ['sub-' cfg.sub], ['ses-' cfg.ses], ['sub-' cfg.sub '_' 'ses-' cfg.ses '_scans.tsv']);
    else
        % construct the output filename, without session directory
        filename = fullfile(cfg.bidsroot, ['sub-' cfg.sub], ['sub-' cfg.sub '_scans.tsv']);
    end

    % get filename
    this = table();
    [p, f, x] = fileparts(cfg.outputfile);
    this.filename = {fullfile(datatype2dirname(cfg.datatype), [f x])};

    fn = fieldnames(cfg.scans);
    for i=1:numel(fn)
        % write [] as 'n/a'
        % write nan as 'n/a'
        % write boolean as 'True' or 'False'
        this.(fn{i}) = output_compatible(cfg.scans.(fn{i}));
    end

    if isfile(filename)
        scans_tsv = ft_read_tsv(filename);
        % the scans.tsv is always merged, use the filename column as the corresponding key
        scans_tsv = mergetable(scans_tsv, this, 'filename');
    else
        scans_tsv = this;
    end

    % the filename should have forward slashes, see #1957 and #1959
    scans_tsv.filename = strrep(scans_tsv.filename, '\', '/');

    % write the updated file back to disk
    ft_write_tsv(filename, scans_tsv);

    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    % update the sessions.tsv, only when sessions are specified
    %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
    if ~isempty(cfg.ses)
        % construct the output filename
        filename = fullfile(cfg.bidsroot, ['sub-' cfg.sub], ['sub-' cfg.sub '_sessions.tsv']);

        % construct session identifier
        this = table();
        session_id = ['ses-' cfg.ses];
        this.session_id = session_id;

        fn = fieldnames(cfg.sessions);
        for i=1:numel(fn)
            % write [] as 'n/a'
            % write nan as 'n/a'
            % write boolean as 'True' or 'False'
            this.(fn{i}) = output_compatible(cfg.sessions.(fn{i}));
        end

        if isfile(filename)
            sessions_tsv = ft_read_tsv(filename);
            % the sessions.tsv is always merged, use the session_id column as the corresponding key
            sessions_tsv = mergetable(sessions_tsv, this, 'session_id');
        else
            sessions_tsv = this;
        end

        % write the updated file back to disk
        ft_write_tsv(filename, sessions_tsv);

    end % if not empty cfg.ses

end % if bidsroot

% do not return an output variable if not requested
if ~ft_nargout
    clear cfg
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function val = get_token(f, tok)
% ensure that it is only the filename
[p, f] = fileparts(f);
pieces = tokenize(f, '_');
val = '';
for i=1:numel(pieces)
    if startsWith([pieces{i} '-'], tok)
        val = pieces{i}((numel(tok)+2):end);
        return
    end
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function f = add_entity(f, key, val)
if ~isempty(val)
    if isscalar(val)
        % this applies specifically to run
        val = num2str(val);
    end
    f = [f '_' key '-' val];
end

function f = remove_entity(f, key)
part = regexp(f, sprintf('_%s-[a-zA-Z0-9+]*', key), 'split');
if numel(part)>1 && ~isempty(part{2})
    f = [part{1} part{2}];
else
    f = part{1};
end

function f = add_datatype(f, typ)
f = [f '_' typ];

function f = remove_datatype(f)
typ = {'FLAIR', 'FLASH', 'PD', 'PDT2', 'PDmap', 'T1map', 'T1rho', 'T1w', 'T2map', 'T2star', 'T2w', 'angio', 'audio', 'bold', 'bval', 'bvec', 'channels', 'coordsystem', 'defacemask', 'dwi', 'eeg', 'emg', 'epi', 'events', 'exg', 'eyetracker', 'fieldmap', 'headshape', 'ieeg', 'inplaneT1', 'inplaneT2', 'magnitude', 'magnitude1', 'magnitude2', 'meg', 'motion', 'nirs', 'phase1', 'phase2', 'phasediff', 'photo', 'physio', 'sbref', 'stim', 'video'};
for i=1:numel(typ)
    if endsWith(f, ['_' typ{i}])
        f = f(1:end-length(typ{i})-1); % also the '_'
        return
    end
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION convert hdr structure into table
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function tab = hdr2table(hdr)
if isempty(hdr)
    % return an empty table
    tab = table();
    return
end

% construct the table, this part applies to all modalities
name = hdr.label(:);
type = hdr.chantype(:);
units = hdr.chanunit(:);
sampling_frequency = repmat(hdr.Fs, hdr.nChans, 1);
tab = table(name, type, units, sampling_frequency);

% NIRS requires some additional columns
if isfield(hdr, 'opto')
    % use the opto structure to determine the source name, detector name and wavelength
    source     = cell(length(name), 1);
    detector   = cell(length(name), 1);
    wavelength = nan(length(name),1); % empty columns will be removed in a later step
    for i=1:length(name)
        labelidx=find(strcmp(hdr.opto.label, name{i}));
        if isempty(labelidx)
            continue
        else
            [dum, optoidx, wavelengthidx]=find(hdr.opto.tra(labelidx,:));
            for k=optoidx
                if any(strcmp(hdr.opto.optotype{k}, {'receiver', 'detector'}))
                    detector{i}=hdr.opto.optolabel{k};
                elseif any(strcmp(hdr.opto.optotype{k}, {'transmitter', 'source'}))
                    source{i}=hdr.opto.optolabel{k};
                end
            end
            if abs(wavelengthidx(1))~= abs(wavelengthidx(2))
                warning('tra matrix is not consistent; ignoring wavelength')
            else
                wavelength(i)=hdr.opto.wavelength(abs(wavelengthidx(1)));
            end
        end
    end

elseif any(strcmpi(hdr.chantype, 'nirs'))
    % deduce the NIRS-specific information from the channel name
    % which typical is something like 'Rx*-Tx* [*wavelength*] or 'S*-D* [*wavelength*]
    source     = cell(length(name), 1);
    detector   = cell(length(name), 1);
    wavelength = nan(length(name),1); % empty columns will be removed in a later step
    for i=1:length(name)
        if regexp(name{i}, 'Rx(\w+)-Tx(\w+) \[(\d+)nm\]')
            parts=regexp(name{i}, 'Rx(?<detectorID>\w+)-Tx(?<sourceID>\w+) \[(?<wavelength>\d+)nm\]', 'names');
            source{i}=sprintf('Tx%s', parts.sourceID);
            detector{i}=sprintf('Rx%s', parts.detectorID);
            wavelength(i)=str2num(parts.wavelength);
        elseif regexp(name{i}, 'Rx(\w+)-Tx(\w+)')
            parts=regexp(name{i}, 'Rx(?<detectorID>\w+)-Tx(?<sourceID>\w+)', 'names');
            source{i}=sprintf('Tx%s', parts.sourceID);
            detector{i}=sprintf('Rx%s', parts.detectorID);
        elseif regexp(name{i}, 'S(\w+)-D(\w+) \[(\d+)nm\]')
            parts=regexp(name{i}, 'S(?sourceID>\w+)-D(?<detectorID>\w+) \[(?<wavelength>\d+)nm\]', 'names');
            source{i}=sprintf('S%s', parts.sourceID);
            detector{i}=sprintf('D%s',parts.detectorID);
            wavelength(i)=str2num(parts.wavelength);
        elseif regexp(name{i}, 'S(\w+)-D(\w+)')
            parts=regexp(name{i}, 'S(?sourceID>\w+)-D(?<detectorID>\w+', 'names');
            source{i}=sprintf('S%s', parts.sourceID);
            detector{i}=sprintf('D%s',parts.detectorID);
        else
            % this channel is not recognized as a nirs channel
            continue
        end
    end
end

if isfield(hdr, 'opto') || any(strcmpi(hdr.chantype, 'nirs'))
    % distinguish between nominal and actual wavelength
    if sum(~isnan(unique(wavelength)))>2
        ft_warning('Assuming that the given wavelengths are actual wavelengths.')
        wavelength_actual = wavelength;
        % try to find the nominal wavelengths:
        split = nanmedian(wavelength);
        WL1.values = wavelength(wavelength<split);
        WL2.values = wavelength(wavelength>split);
        WL1.nominal = round(median(WL1.values),-1);
        WL2.nominal = round(median(WL2.values),-1);
        ft_warning('assuming that the nominal wavelengths are %d and %d nm', WL1.nominal, WL2.nominal)
        wavelength_nominal = nan(size(wavelength));
        for i=1:length(wavelength)
            if any(wavelength(i)==WL1.values)
                wavelength_nominal(i) = WL1.nominal;
            elseif any(wavelength(i)==WL2.values)
                wavelength_nominal(i) = WL2.nominal;
            end
        end
    else
        ft_warning('Assuming that the given wavelengths are nominal wavelengths.')
        wavelength_nominal = wavelength;
        wavelength_actual = nan(size(wavelength));
    end

    % add the source, detector and wavelength columns to the table
    tab = horzcat(tab, table(source, detector, wavelength_nominal, wavelength_actual));
end


%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION convert elec structure into table
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function tab = elec2table(elec)
if isempty(elec)
    tab = table();
else
    name = elec.label(:);
    x = elec.elecpos(:,1);
    y = elec.elecpos(:,2);
    z = elec.elecpos(:,3);
    tab = table(name, x, y, z);
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION convert opto structure into table
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function tab = opto2table(opto)
if isempty(opto)
    tab = table();
else
    name = opto.optolabel(:);
    type = cell(size(name));
    idx_source = find(contains(opto.optotype, {'transmitter', 'source'}));
    idx_detector = find(contains(opto.optotype, {'receiver', 'detector'}));
    type(idx_source) = {'source'};
    type(idx_detector) = {'detector'};
    if all(opto.optopos(:,3)==0) % these are probably template positions
        ft_info('assuming the optode positions are template positions');
        x=nan(length(name),1);
        y=nan(length(name),1);
        z=nan(length(name),1);
        template_x=opto.optopos(:,1);
        template_y=opto.optopos(:,2);
        template_z=opto.optopos(:,3);
    else % these are probably recorded optode positions
        ft_info('assuming the optodes are recorded positions')
        x = opto.optopos(:,1);
        y = opto.optopos(:,2);
        z = opto.optopos(:,3);
        template_x=nan(length(name),1);
        template_y=nan(length(name),1);
        template_z=nan(length(name),1);
    end
    tab = table(name, type, x, y, z, template_x, template_y,template_z);
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION convert event structure into table
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function tab = event2table(hdr, event)
if isempty(event)
    tab = table();
else
    for i=1:numel(event)
        % normalize missing values
        if isempty(event(i).sample)
            event(i).sample = NaN;
        end
        if isempty(event(i).duration)
            event(i).duration = NaN;
        end
        if isempty(event(i).value)
            event(i).value = NaN;
        end
    end
    if isempty(hdr) && isfield(event, 'timestamp')
        ft_warning('assuming a timestamp resolution of 0.1 milliseconds');
        onset        = ([event.timestamp]/10000)';     % the resolution is 0.1 milliseconds
        duration     = zeros(size(onset));             % in seconds
        type         = {event.type}';
        value        = {event.value}';
    else
        onset        = (([event.sample]-1)/hdr.Fs)';   % in seconds
        duration     = ([event.duration]/hdr.Fs)';     % in seconds
        sample       = ([event.sample])';              % in samples, the first sample of the file is 1
        type         = {event.type}';
        value        = {event.value}';
    end
    if all(cellfun(@isnumeric, type))
        % this can be an array of strings or values
        type = cell2mat(type);
    end
    if all(cellfun(@isnumeric, value))
        % this can be an array of strings or values
        value = cell2mat(value);
    end
    if exist('sample', 'var')
        tab = table(onset, duration, sample, type, value);
    else
        tab = table(onset, duration, type, value);
    end
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function filename = corresponding_json(filename)
[p, f, x] = fileparts(filename);
if isequal(x, '.gz') && endsWith(f, '.nii')
    % it is a gzip compressed NIfTI file, remove the .nii from the file name
    f = f(1:end-4);
end
filename = fullfile(p, [f '.json']);

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function x = merge_vector(x, y)
assert(isequal(size(x), size(y)));
for i=1:numel(x)
    if isnumeric(x) && isnumeric(y) && isnan(x(i)) && ~isnan(y(i))
        x(i) = y(i);
    end
    if iscell(x) && iscell(y) && isempty(x{i}) && ~isempty(y{i})
        x{i} = y{i};
    end
    if iscell(x) && isnumeric(y) && isempty(x{i}) && ~isnan(y{i})
        x{i} = y(i);
    end
    if iscell(x) && iscell(y) && isequal(x{i}, 'unknown') && ~isequal(y{i}, 'unknown')
        x{i} = y{i};
    end
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function sel = select_event(event, eventtype, eventvalue)
% this code is copied from FT_TRIALFUN_GENERAL

% start by selecting all events
sel = true(1, length(event)); % this should be a row vector

% select all events of the specified type
if ~isempty(eventtype)
    for i=1:numel(event)
        sel(i) = sel(i) && ismatch(event(i).type, eventtype);
    end
end

% select all events with the specified value
if ~isempty(eventvalue)
    for i=1:numel(event)
        sel(i) = sel(i) && ismatch(event(i).value, eventvalue);
    end
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function s = remove_empty(s)
if isempty(s)
    return
elseif isstruct(s)
    fn = fieldnames(s);
    fn = fn(structfun(@isempty, s));
    s = removefields(s, fn);
elseif istable(s)
    remove = false(1,size(s,2));
    for i=1:size(s,2)
        % find columns that are non-numeric and where all elements are []
        remove(i) = ~isnumeric(s{:,i}) && all(cellfun(@isempty, s{:,i}));
    end
    s = s(:,~remove);
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function y = sort_columns(x, desired)
original = x.Properties.VariableNames;
if nargin<2
    % sort alphabetically
    desired = sort(original);
else
    % only keep the desired columns that are actually present
    desired = intersect(desired, original, 'stable');
end
[dum, order] = ismember(desired, original);
y = x(:, order);

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function y = sort_fields(x, desired)
original = fieldnames(x);
if nargin<2
    % sort alphabetically
    desired = sort(original);
else
    % only keep the desired fields that are actually present
    desired = intersect(desired, original, 'stable');
end
y = struct();
for i=1:numel(desired)
    y.(desired{i}) = x.(desired{i});
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function dir = datatype2dirname(typ)
% see https://bids-specification.readthedocs.io/en/stable/99-appendices/04-entity-table.html
% nirs, emg, exg, eyetracker, motion, audio, and video are not part of the official specification
switch typ
    case {'T1w' 'T2w' 'T1rho' 'T1map' 'T2map' 'T2star' 'FLAIR' 'FLASH' 'PD' 'PDmap' 'PDT2' 'inplaneT1' 'inplaneT2' 'angio' 'defacemask'}
        dir = 'anat';
    case {'bold' 'sbref'} % this could also include 'events' or other non-brain data
        dir = 'func';
    case {'dwi' 'bvec' 'bval'}
        dir = 'dwi';
    case {'phasediff' 'phase1' 'phase2' 'magnitude1' 'magnitude2' 'magnitude' 'fieldmap' 'epi'}
        dir = 'fmap';
    case {'events' 'stim' 'physio' 'emg' 'exg' 'eyetracker' 'audio' 'video'} % these could also all be stored in 'func' or one of the other directories with brain data
        dir = 'beh';
    case {'meg'} % this could also include 'events' or other non-brain data
        dir = 'meg';
    case {'eeg'} % this could also include 'events' or other non-brain data
        dir = 'eeg';
    case {'ieeg'} % this could also include 'events' or other non-brain data
        dir = 'ieeg';
    case {'motion'} % this is not part of the official specification
        dir = 'motion';
    case {'nirs'} % this is not part of the official specification
        dir = 'nirs';

    otherwise
        ft_error('unrecognized data type ''%s''', typ);
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function val = output_compatible(val)
if istable(val)
    fn = val.Properties.VariableNames;
    for i=1:numel(fn)
        val.(fn{i}) = output_compatible(val.(fn{i}));
    end
elseif iscell(val)
    % use recursion to make all elements compatible
    val = cellfun(@output_compatible, val, 'UniformOutput', false);
elseif isnumeric(val) && numel(val)>1 && any(isnan(val))
    % convert and use recursion to make all elements compatible
    val = num2cell(val);
    val = cellfun(@output_compatible, val, 'UniformOutput', false);
elseif isdatetime(val)
    % see https://bids-specification.readthedocs.io/en/stable/02-common-principles.html#units
    val = datestr(val, 'yyyy-mm-ddTHH:MM:SS');
else
    % write [] as 'n/a'
    % write nan as 'n/a'
    % write boolean as 'True' or 'False'
    if isempty(val)
        val = 'n/a';
    elseif isnan(val)
        val = 'n/a';
    elseif islogical(val)
        if val
            val = 'True';
        else
            val = 'False';
        end
    end
end

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% SUBFUNCTION
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
function t = convert_table(s)
assert(isstruct(s));
assert(numel(s)==1);
fn = fieldnames(s);
for i=1:numel(fn)
    if ischar(s.(fn{i}))
        % convert to cell-array
        s.(fn{i}) = {s.(fn{i})};
    elseif iscell(s.(fn{i}))
        % ensure it is a column
        s.(fn{i}) = s.(fn{i})(:);
    elseif isnumeric(s.(fn{i}))
        % ensure it is a column
        s.(fn{i}) = s.(fn{i})(:);
    end
end
t = struct2table(s);