artifacts: Generate some artifacts for a multi-dimensional image-like...

In ygu427/iSPREADR: imProved Spatial Regression Analysis for DTI Data in R

Description

Generate multivariate Gaussian artifacts for a multi-dimensional image-like data (array object).

Usage

artifacts(grids, center=rep(0, length(grids)), std=0.3, radius=1.0, magnitude=1.0, xscale=diag(length(grids)))

Arguments

grids	A list of monotonically increasing coordinate grids. For example, grids <- list(grid.x=seq(0, 100, 2), grid.y=exp(seq(1, 5, .2))).
center	The center of the artifact. Default value: the origin. Example: center=c(3,5) defines a center for 2d data.
std	The standard deviation of the (unscaled) Gaussian density is set to std*radius. Default value: 0.5.
radius	The radius of the (unscaled) artifact. Generated value outside of this radius is zero. Default to 10.
maxval	Generated Gaussian artifacts will be scaled so that the maxium value equals maxval. Default to 1.0
minval	Generated Gaussian artifacts will be set to zero if they are less thanminval. Default to 0.2.
xscale	A scaling matrix for the shape of the artifact. Use this to generate elliptic artifacts. Default to the identity matrix.

Details

The unscaled artifacts is a multivariate Gaussian density centered at center with standard deviation std. Its support is a ball of radius radius. A user specified xscale matrix can be used to stretch/rotate this ball.

Value

An array defined on grids which represents an artifact. It can be added to an existing array with the same grids.

Author(s)

Xing Qiu

Examples

## A3 is a 3x3 scaling matrix.
A3 <- matrix(c(1.5, 0.5, 0.0,
               0.5, 1.5, 0.0,
               0.0, 0.0, 1.0), nrow=3)
## From a DTI related simulation, 128x128x3 voxels, each 2x2x2mm large.
grid.x <- seq(1, 255, 2); grid.y <- seq(1, 255, 2); grid.z <- seq(1,5,2)
N.x <- length(grid.x); N.y <- length(grid.y); N.z <- length(grid.z)
grids <- list(grid.x, grid.y, grid.z)

##Sharp effects,  60 voxels, Max: 10. 
Z1 <- artifacts(grids, center=c(50,80,3), xscale=A3, radius=4, maxval=10.0, minval=3) 

## Diffused effects, 516 voxels. Max: 1.2.
Z2 <- artifacts(grids, center=c(50,80,3), maxval=1.2, radius=15)

ygu427/iSPREADR documentation built on May 20, 2019, 4:37 p.m.