Movatterモバイル変換

fMRI timeseries data in CIFTI format ("*.dtseries.nii").For single-session analysis this can be a file path to a CIFTI file or a"xifti" object from theciftiTools package. For multi-sessionanalysis this can be a vector of file paths or a list of"xifti"objects.

IfBOLD is a"xifti" object(s), the surfaces, if any, will beused for the spatial model. However, ifsurfL andsurfR areprovided, they will override any surfaces inBOLD.

SeeBayesGLM.

SeeBayesGLM.

SeeBayesGLM.

SeeBayesGLM.

SeeBayesGLM.

SeeBayesGLM.

SeeBayesGLM.

BOLD, design, nuisance

(Optional) canonical design matrix

verbose?

time vector

0 (default) for the HRF,1 for the first derivativeof the HRF, or2 for the second derivative of the HRF.

delay of response. Default:6

response dispersion. Default:0.9

delay of undershoot. Default:12

dispersion of undershoot. Default:0.9

scale of undershoot. Default:0.35

time vector (in units of seconds)

0 (default) for the HRF,1 for the delay derivativeof the HRF, or2 for the dispersion derivative of the HRF.

delay of response. Default:6

response dispersion. Default:1

delay of undershoot. Default:16/6 * a1 * sqrt(b1) = 16

dispersion of undershoot. Default:b1 = 1

scale of undershoot. Default:1/6

onset of response. Default:0

time vector (in seconds). Must be equally spaced.

delay of response. Default:6

response dispersion. Default:1

delay of undershoot. Default:16/6*a1 = 16

dispersion of undershoot. Default:b1 = 1

scale of undershoot. Default:1/6

onset of response (in seconds). Default:0

Temporal resolution of the data, in seconds.

(For prewhitening) Use the Akaike information criterion (AIC) toselect AR model orders between0 andar_order? Default:FALSE.

(For prewhitening) The order of the autoregressive (AR) modelto use for prewhitening. If0, do not prewhiten. Default:6.

For multi-session modeling, note that a single AR model is used; itscoefficients will be the average estimate from each session.

(For prewhitening) The FWHM parameter for spatiallysmoothing the coefficient estimates for the AR model to use forprewhitening. Recall that\sigma = \frac{FWHM}{2*sqrt(2*log(2)}. Set to0 to not smooththe estimates. Default:5.

Character vector indicating which brain structure(s)ofBOLD to analyze:"left" cortex;"right" cortex;and/or"subcortical" structures. Or"all" to model all three.Default:c("left","right") (cortex only).

NULL or a numeric matrix

A numeric matrix with the same number of rows asmat_or_NULL

A numeric matrix, or a vector which will be converted to asingle-column matrix.

A numeric matrix ordata.frame, or a"BayesfMRI_design" object frommake_design. Can alsobe an array where the third dimension is the same length as the number ofdata locations, to model each location with its own design.

A session-length list ofT \times V numeric BOLD data.

Tolerance for mean and variance of each data location.Locations which do not meet these thresholds are masked out of the analysis.Defaults:1e-6.

Print messages counting how many locations are removed?Default:TRUE.

AnF \times 3 matrix, where each row contains the vertexindices for a given triangular face in the mesh.F is the number offaces in the mesh.

(Optional) Names of fields represented in design matrix.

The putative EVs matrix or data frame

Thedesign argument input. Will be formatted to anS-length list.

Scale the design matrix by dividing each column by itsmaximum and then subtracting the mean? Default:TRUE. IfFALSE, the design matrix is centered but not scaled.

The expected number of sessions, if known.

The expected number of timepoints, if known. Formulti-session data this is a session-length vector.

The expected number of designs, if known. For per-locationmodeling this is equal tonVd0, the initial number of data locations.For multi-session data this is a session-length vector.

FALSE if per-location modeling is notbeing performed (i.e. for multiGLM);TRUE if it is; or,NULLto infer based on the dimensions ofdesign (TRUE if thedesign has three dimensions.)

Thenuisance argument input. Will be formatted to anS-length list.

The expected number of sessions, if known.

The expected number of timepoints, if known. Formulti-session data this is a session-length vector.

Thescrub argument input. Will be formatted to anS-length list.

The expected number of sessions, if known.

The expected number of timepoints, if known. Formulti-session data this is a session-length vector.

Add DCT bases tonuisance to apply a temporal high-passfilter to the data, for detrending?hpf is the filter frequency.UseNULL to skip detrending. Detrending is strongly recommended forfMRI data, to help reduce the autocorrelation in the residuals, soNULL will induce a warning. Use"already" to disable thewarning while skipping highpass filtering.

Using at least two DCT bases is as sufficient for detrending as using linearand quadratic drift terms in the nuisance matrix. So if DCT detrending isbeing used here, there is no need to add linear and quadratic drift terms tonuisance.

The object

The design matrix/array

The nuisance matrix

The scrub matrix

The explanatory variables i.e. the task stimulus information,from which a design matrix will be constructed. This is a list where eachentry represents a task as a matrix of onsets (first column) and durations(second column) for each stimuli (each row) of the task, in seconds. Listnames should be the task names.nTime andTR are required.

An example of a properly-formattedEVs is:on_s1 <- list(taskA=cbind(on=c(1,9,17), dr=rep(1,3)), taskB=cbind(on=c(3,27), dr=rep(5,2))).In this example, there are two tasks: the first has three 1s-long stimuli,while the second has two 5s-long stimuli.

the number of timepoints (volumes) in the task fMRI data.

the temporal resolution of the data, in seconds.

Controls the extent of HRF derivatives modeling.

Set to0 to only model the main HRF regressor (default), and not include itsderivatives; set to1 to model the temporal derivative too;or, set to2 to model both the temporal and dispersion derivatives.IfdHRF==0, there is one design column (field) per task. IfdHRF==1, there are two fields per task. And ifdHRF==2, thereare three fields per task.

If there are several tasks anddHRF>0, the total number of designmatrix columns may exceed five, which may require large computation timeswith INLA. The analysis can be adjusted by modeling the derivatives asnuisance signals rather than as fields. To do so, move the correspondingcolumns from the design matrix to thenuisance argument forBayesGLM.

Upsample factor for convolving stimulus boxcar or stickfunction with canonical HRF. Default:100.

Add task regressors indicating the onset and/or offset ofeach event block? Provide the names of the tasks as a character vector. Allonsets (or offsets) across the specified tasks will be represented by oneadditional column in the design matrix. The task names must match the namesofEVs. Can also be"all" to use all tasks.

Onsets/offset modeling is only compatible with a block design experiment.An error will be raised if the events inEVs do not have durationgreater than one second.

Scale the columns of the design matrix? Default:TRUE.

Model the onsets (onsets_sep) or offsets(offsets_sep) separately for each task? Default:FALSE, tomodel all onsets together, or all offsets together, as a single field in thedesign.

Print diagnostic messages? Default:TRUE.

Additional arguments toHRF_calc.

A lengthV logical vector indicating if each vertex iswithin the input mask.

Tolerance for mean and variance of each data location.Locations which do not meet these thresholds are masked out of the analysis.Default:1e-6 for both.

fMRI timeseries data in CIFTI format ("*.dtseries.nii").For single-session analysis this can be a file path to a CIFTI file or a"xifti" object from theciftiTools package. For multi-sessionanalysis this can be a vector of file paths or a list of"xifti"objects.

IfBOLD is a"xifti" object(s), the surfaces, if any, will beused for the spatial model. However, ifsurfL andsurfR areprovided, they will override any surfaces inBOLD.

A 3D numeric array that is locations by fields by designs.

Character vector indicating which brain structure(s)ofBOLD to analyze:"left" cortex;"right" cortex;and/or"subcortical" structures. Or"all" to model all three.Default:c("left","right") (cortex only).

Temporal resolution of the data, in seconds.

For cortex spatial model. The number of vertices to whicheach cortical surface should be resampled, orNULL to not resample.

For computational feasibility, a value of10000 (default) or lower isrecommended for Bayesian spatial modeling. IfBayes=FALSE,resamp_res can be set toNULL for full-resolution classicalmodeling.

Add DCT bases tonuisance to apply a temporal high-passfilter to the data, for detrending?hpf is the filter frequency.UseNULL to skip detrending. Detrending is strongly recommended forfMRI data, to help reduce the autocorrelation in the residuals, soNULL will induce a warning. Use"already" to disable thewarning while skipping highpass filtering.

Using at least two DCT bases is as sufficient for detrending as using linearand quadratic drift terms in the nuisance matrix. So if DCT detrending isbeing used here, there is no need to add linear and quadratic drift terms tonuisance.

(Optional) AT \times N_{nuis} matrix of nuisance signals,whereT is the number of timepoints andN is the number ofnuisance signals, or a list of these for multi-session analysis. Nuisancesignals are regressed from the fMRI data and design matrix prior to GLMcomputation. Nuisance signals can include motion regressors, HRF derivativesnot being modeled as tasks, and other sources of noise.

Detrending/high-pass filtering is accomplished by adding DCT bases to thenuisance matrix; see the parametershpf andDCT.

Do not add spike regressors for scrubbing to thenuisance matrix.Rather, provide these inscrub so that their corresponding timepointsare also removed from the BOLD data after nuisance regression.

TO DO

1 (default) to print occasional updates during modelcomputation;2 for occasional updates as well as running INLA inverbose mode (ifBayes), or0 for no printed updates.

Tolerance for mean and variance of each data location.Locations which do not meet these thresholds are masked out of the analysis.Default:1e-6 for both.

Session-length list of numeric matrices/arrays,each with volumes along the first dimension.

TO DO

1 (default) to print occasional updates during modelcomputation;2 for occasional updates as well as running INLA inverbose mode (ifBayes), or0 for no printed updates.

Tolerance for mean, variance and SNR of each data location.Locations which do not meet these thresholds are masked out of the analysis.Default:1e-6 for mean and variance,50 for SNR.

(Optional) AT \times N_{nuis} matrix of nuisance signals,whereT is the number of timepoints andN is the number ofnuisance signals, or a list of these for multi-session analysis. Nuisancesignals are regressed from the fMRI data and design matrix prior to GLMcomputation. Nuisance signals can include motion regressors, HRF derivativesnot being modeled as tasks, and other sources of noise.

Detrending/high-pass filtering is accomplished by adding DCT bases to thenuisance matrix; see the parametershpf andDCT.

Do not add spike regressors for scrubbing to thenuisance matrix.Rather, provide these inscrub so that their corresponding timepointsare also removed from the BOLD data after nuisance regression.

An object of class "BfMRI_design".

Additional arguments toplot_design.

The timepoints by fields design matrix or data.frame.

"lineplot" (default) or"imageplot".

Additional arguments toplot_design_line orplot_design_image.

The name of a ColorBrewer palette (seeRColorBrewer::brewer.pal.info and colorbrewer2.org), the name of aviridisLite palette, or a character vector of colors. Default:"Set1".

Parameters forggplot2::geom_line.Defaults:"solid" linetype,0.7 linewidth and0.8alpha.linetype can also be a vector of options with length matchingthe number of fields indesign.

For cortex spatial model. The number of vertices to whicheach cortical surface should be resampled, orNULL to not resample.

For computational feasibility, a value of10000 (default) or lower isrecommended for Bayesian spatial modeling. IfBayes=FALSE,resamp_res can be set toNULL for full-resolution classicalmodeling.

Controls scaling the BOLD response at each location.

"mean":

Scale the data to percent local signal change.

"sd":

Scale the data by local standard deviation.

"none":

Center the data but do not scale it.

The original (unscaled) design matrix that is T x K, whereT is the number of time points, and k is the number of field covariates

(Optional) AT \times N_{scrub} matrix of spike regressors(one 1 value at the timepoint to scrub, and 0 for all other values), or alogical vector indicating the timepoints to scrub (TRUE to scrub, andFALSE to keep). For multi-session data, a session-length list ofsuch matrices or logical vectors.

The spike regressors will be included in the nuisanceregression, and afterwards the timepoints indicated inscrub will beremoved from the BOLD data and design matrix.

The names of the task-fMRIBOLD sessions, formulti-session analysis. If not provided here, will be inferred fromnames(BOLD), inferred fromnames(design), or generatedautomatically, in that order.

Object of class"BfMRI_design".

further arguments passed to or from other methods.

Object of class"summary.BfMRI_design".

For cortex spatial model. Left and right cortex surfacegeometry in GIFTI format ("*.surf.gii"). These can be a file path toa GIFTI file or a"surf" object fromciftiTools.

Surfaces can alternatively be provided through the$surf metadata inBOLD if it is"xifti" data. If neither are provided, by default theHCP group-average fs_LR inflated surfaces included inciftiTools will beused for the cortex spatial model.

1 (default) to print occasional updates during modelcomputation;2 for occasional updates as well as running INLA inverbose mode (ifBayes), or0 for no printed updates.