Note that this reference documentation is identical to the help that is displayed in MATLAB when you type “help ft_rejectartifact”.
FT_REJECTARTIFACT removes data segments containing artifacts. It returns a configuration structure with a modified trial definition which can be used for preprocessing of only the clean data. You should start by detecting the artifacts in the data using the function FT_ARTIFACT_xxx where xxx is the type of artifact. Subsequently FT_REJECTARTIFACT looks at the detected artifacts and removes them from the trial definition or from the data. In case you wish to replace bad parts by nans, you have to specify data as an input parameter. Use as cfg = ft_rejectartifact(cfg) with the cfg as obtained from FT_DEFINETRIAL, or as data = ft_rejectartifact(cfg, data) with the data as obtained from FT_PREPROCESSING The following configuration options are supported: cfg.artfctdef.reject = 'none', 'partial','nan', or 'complete' (default = 'complete') cfg.artfctdef.minaccepttim = when using partial rejection, minimum length in seconds of remaining trial (default = 0.1) cfg.artfctdef.crittoilim = when using complete rejection, reject trial only when artifacts occur within this time window (default = whole trial) (only works with in-memory data, since trial time axes are unknown for data on disk) cfg.artfctdef.feedback = 'yes' or 'no' (default = 'no') cfg.artfctdef.eog.artifact = Nx2 matrix with artifact segments, this is added to the cfg by using FT_ARTIFACT_EOG cfg.artfctdef.jump.artifact = Nx2 matrix with artifact segments, this is added to the cfg by using FT_ARTIFACT_JUMP cfg.artfctdef.muscle.artifact = Nx2 matrix with artifact segments, this is added to the cfg by using FT_ARTIFACT_MUSCLE cfg.artfctdef.zvalue.artifact = Nx2 matrix with artifact segments, this is added to the cfg by using FT_ARTIFACT_ZVALUE cfg.artfctdef.xxx.artifact = Nx2 matrix with artifact segments, this should be added by your own artifact detection function A trial that contains an artifact can be rejected completely or partially. In case of partial rejection, a minimum length of the resulting sub-trials can be specified. Output: If cfg is used as the only input parameter, a cfg with a new trl is the output. If cfg and data are both input parameters, a new raw data structure with only the clean data segments is the output. To facilitate data-handling and distributed computing you can use cfg.inputfile = ... If you specify this option the input data will be read from a *.mat file on disk. This mat files should contain only a single variable named 'data', corresponding to the input structure. See also FT_ARTIFACT_EOG, FT_ARTIFACT_MUSCLE, FT_ARTIFACT_JUMP, FT_ARTIFACT_MANUAL, FT_ARTIFACT_THRESHOLD, FT_ARTIFACT_CLIP, FT_ARTIFACT_ECG