simTSrestingstate: Simulate fMRI resting state time series

Description

Synthesizes a single time series x representing resting state activity. The fluctuation frequencies f are limited to a square passband 0.01 Hz <= f <= 0.1 Hz. TR is the repetition time (needed to compute the passband limits), expressed in seconds. N is the required number of samples (needs not be a power of 2).

Usage

simTSrestingstate(nscan, base=0, TR, SNR=NULL, noise = c("none", "white",
	"temporal", "low-frequency", "physiological", "mixture"), 
	type = c("gaussian", "rician"), weights, verbose = TRUE, rho = 0.2,
	freq.low = 128, freq.heart = 1.17, freq.resp = 0.2, vee=1)

Value

A vector representing the resting state time series

Arguments

nscan: Number of scans.
base: Baseline value of the time series.
TR: Repetition time in seconds.
SNR: Signal-to-noise ratio of the time series.
noise: Type of noise (white is default).
type: If noise==white, noise==mixture, type of system noise (gaussian is default).
weights: If noise==mixture, vector of weights to specify the fraction of the noise components.
verbose: Logical indicating if warnings should be returned.
rho: If noise==temporal or noise==mixture, value of autocorrelation coefficients. The length of the vector corresponds to the order of the autoregressive model.
freq.low: If noise==low-frequency or noise==mixture, frequency of the low-frequency drift in seconds.
freq.heart: If noise==physiological or noise==mixture, frequency of heart rate in Hz.
freq.resp: If noise==physiological or noise==mixture, frequency of respiratory rate in Hz.
vee: If type==rician, non-centrality parameter of the distribution.

Author

J. Durnez, G. Verdoolaege, M. Welvaert

References

[1] C.G. Fox, Computers & Geoscience, Vol. 13, pp. 369-374, 1987.

[2] M. Fukunaga, Magnetic Resonance Imaging, Vol. 24, pp. 979-992, 2006.

Examples

Run this code

out <- simTSrestingstate(nscan=50, TR=2, SNR=1, noise="none")
plot(out, type="l")
rm(out)

Run the code above in your browser using DataLab