This function prepares an fMRI linear model for AFNI's 3dDeconvolve tool. It takes an fmri_model object, an fmri_dataset object, and various options to control the fitting process.
Usage
afni_lm(
fmri_mod,
dataset,
working_dir = ".",
polort = -1,
jobs = 1,
censor = NULL,
options = list()
)Arguments
- fmri_mod
An fmri_model object containing the event and baseline models
- dataset
An fmri_dataset object containing the scan data and other necessary information
- working_dir
The working directory (default is the current directory)
- polort
The number of polynomial baseline regressors (default is to suppress 'polort')
- jobs
The number of jobs to use with '3dDeconvolve' (default is 1)
- censor
A list of censoring vectors, one per run, or a single vector equal to the total number of scans (default is NULL)
- options
A list of options to be sent to 3dDeconvolve (default is an empty list)
Value
An afni_lm_spec object containing the fitted model, dataset, working directory, options, and command
Examples
etab <- data.frame(onset=c(1,30,15,25), fac=factor(c("A", "B", "A", "B")),
run=c(1,1,2,2))
dset <- fmri_dataset(scans=c("s1.nii", "s2.nii"), mask="mask.nii", TR=1,
run_length=c(50,50), event_table=etab)
emodel <- event_model(onset ~ hrf(fac), block = ~ run, data=etab,
sampling_frame=dset$sampling_frame)
#> [1] "onset" "hrf(fac)"
bmodel <- baseline_model("bs", degree=4, sframe=dset$sampling_frame)
fmod <- fmri_model(emodel, bmodel)
alm <- afni_lm(fmod, dset, jobs=2, options=list(tout=TRUE, errts="residuals.nii.gz"))