arf: Activated Region Fitting
In arfS4: Activated Region Fitting, fMRI data analysis.

Description Usage Arguments Details Value Author(s) References

A call to arf starts the ARF fitting procedure. It checks if the given path contains a "data" and a "weights" directory and creates the necessary S4 classes and average files (if enabled). The name of the experiment can be entered via expname, this is only used for output on screen. The sequence and regionstofit parameters define the number of regions that are fitted and in which order (see the arguments section for details). The fast option enables/disables calculation of sandwich estimates at each model that is fitted or only at the best fitting model. fullOutput controls the amount of output shown on screen. createAverages controls the creation of the average files.

1	arf(path = "", expname = NULL, sequence = c("manual", "sequential", "interleaved"), regionstofit = 1, fast = F, fullOutput = T, createAverages = T)

`path`	The directory containing the data (it assumes the "data" and "weights" directories exist in the given path). See `makeDirStruct` on how to automatically create a proper directory structure.
`expname`	The name of the experiment/condition
`sequence`	Can be either `manual`, `sequential`, or `interleaved`.
`regionstofit`	With sequence set to 'manual': a vector containing the sequence of models to fit (values indicate the number of regions to fit). With sequence set to 'sequential': an integer giving the maximum number of regions to fit. ARF then fits all models from 1 to the maximum. With sequence set to 'interleaved': an integer giving the maximum number of regions to fit. ARF then fits all odd models from 1 to the maximum.
`fast`	When fast is `FALSE` (default) the sandwich estimates are calculated at each model in the sequence. When fast is `TRUE` sandwich estimates are only calculated for the best fitting model.
`fullOutput`	When fullOutput is `TRUE` (default) all output is displayed, when `FALSE` only selected output is given.
`createAverages`	When createAverages is `TRUE` (default) averages are calculated, when `FALSE` they are not (and are assumed to exist in the 'avg' directory.

ARF can read in NIFTI formatted files or analyze files, assuming a neurological orientation. The NIFTI/ANALYZE header parameters (file type etc.) are taken from the first file in the 'data' directory. For read/write related functions see getFileInfo,c odenewFile, codereadData, codereadHeader, codewriteData, codewriteHeader.

none. The S4 object of class data of each model can be accessed via the .RDA files in the 'stats' directory. The S4 object of class sequence containing the model fits can also be accessed via this directory. See codedata and codesequence for details on the S4 classes.

Wouter D.Weeda

Weeda, W.D., Waldorp, L.J., Christoffels, I., and Huizenga, H.M. (in press) Activated Region Fitting: A robust high-power method for fMRI analysis using parameterized regions of activation. Human Brain Mapping, 2009.

arfS4 documentation built on May 2, 2019, 6:14 p.m.