Working with Image Volumes

In neuroim: Data Structures and Handling for Neuroimaging Data

knitr::opts_chunk$set(collapse = T, comment = "#>")
library(neuroim)

Reading a NIFTI formatted image volume

The easiest way to read an image file is to use loadVolume:

    fileName <- system.file("extdata", "global_mask.nii", package="neuroim")
    vol <- loadVolume(fileName)

Working with image volumes

Information about the geometry of the image volume is easily accessed:

    print(vol)

loadVolume returns an object of class DenseBrainVolume which extends an R `array' and has 3 dimensions (x,y,z).

    class(vol)

    is.array(vol)

    dim(vol)

    vol[1,1,1]

    vol[64,64,24]

Arithmetic can be performed on images as if they were ordinary arrays:

    vol2 <- vol + vol
    sum(vol2) == 2 * sum(vol)

    vol3 <- vol2 - 2*vol
    all(vol3 == 0)

A numeric image volume can be converted to a binary image as follows:

    vol2 <- as.logical(vol)
    print(vol2[1,1,1])

We can also create a BrainVolume instance from an array or numeric vector:

    # create an 64X64X64 array of zeros 
    x <- array(0, c(64,64,64))

    # create a 'BrainSpace' instance that describes the geometry of the image including, at minimu its dimensions and voxel spacing
    bspace <- BrainSpace(Dim=c(64,64,64), spacing=c(1,1,1))
    vol <- BrainVolume(x, bspace)
    vol

We do not usually have to create BrainSpace objects because this information is usually read from disk. Thus, BrainSpace objects are usually copied from existing images using the space extractor function when needed:

    vol2 <- BrainVolume((vol+1)*25, space(vol))
    max(vol2)

    space(vol2)

Writing a NIFTI formatted image volume

When we're ready to write an image volume to disk, we use writeVolume

    writeVolume(vol2, "output.nii")

    ## adding a '.gz' extension results ina gzipped file.
    writeVolume(vol2, "output.nii.gz")

neuroim documentation built on May 2, 2019, 1:04 p.m.