ANTsR (version 0.3.3)

jointLabelFusion: joint label fusion

Description

A multiple atlas voting scheme to customize labels for a new subject.

Usage

jointLabelFusion(targetI, targetIMask, atlasList, beta = 4, rad = NA, labelList = NA, doscale = TRUE, doNormalize = TRUE, maxAtlasAtVoxel = c(1, Inf), rho = 0.01, usecor = FALSE, boundary.condition = "image", rSearch = 2, segvals = NA, probimgs = NA, jifImage = NA)

Arguments

targetI

antsImage to be approximated

targetIMask

mask with value 1

atlasList

list containing antsImages

beta

weight sharpness, default to 2

rad

neighborhood radius, default to 4

labelList

list containing antsImages

doscale

scale neighborhood intensities

doNormalize

normalize each image range to 0, 1

maxAtlasAtVoxel

min/max n atlases to use at each voxel

rho

ridge penalty increases robustness to outliers but also makes image converge to average

usecor

employ correlation as local similarity

boundary.condition

one of 'image' 'mean' 'NA'

rSearch

radius of search, default is 2

segvals

list of labels to expect

probimgs

list of probability images (to help slice by slice estimate)

jifImage

the current estimated jif image (helps speed slice by slice)

Value

approximated image, segmentation and probabilities

Examples

set.seed(123)
ref<-antsImageRead( getANTsRData("r16"))
ref<-resampleImage(ref,c(50,50),1,0)
ref<-iMath(ref,"Normalize")
mi<-antsImageRead( getANTsRData("r27"))
mi2<-antsImageRead( getANTsRData("r30"))
mi3<-antsImageRead( getANTsRData("r62"))
mi4<-antsImageRead( getANTsRData("r64"))
mi5<-antsImageRead( getANTsRData("r85"))
refmask<-getMask(ref)
refmask<-iMath(refmask,"ME",2) # just to speed things up
ilist<-list(mi,mi2,mi3,mi4,mi5)
seglist<-list()
for ( i in 1:length(ilist) )
 {
 ilist[[i]]<-iMath(ilist[[i]],"Normalize")
 mytx<-antsRegistration(fixed=ref , moving=ilist[[i]] ,
   typeofTransform = c("Affine") )
 mywarpedimage<-antsApplyTransforms(fixed=ref,moving=ilist[[i]],
   transformlist=mytx$fwdtransforms)
 ilist[[i]]=mywarpedimage
 seg<-kmeansSegmentation( ilist[[i]], k=3, kmask = refmask)
 seglist[[i]]<-seg$segmentation
 }
r<-2
pp<-jointLabelFusion(ref,refmask,ilist, rSearch=2,
  labelList=seglist, rad=rep(r, length(dim(ref)) ) )