-
-
Notifications
You must be signed in to change notification settings - Fork 387
/
Copy pathantsASLProcessing.R
executable file
·281 lines (259 loc) · 11.4 KB
/
antsASLProcessing.R
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
#!/usr/bin/env Rscript
library(ANTsR)
library(tools)
if(!usePkg('optparse') | !usePkg('ANTsR')){
stop("optparse and ANTsR packages required.")
}
optlist <- list(
make_option(c('-s', '--pCASL'), default='', help=' raw pCASL image'),
make_option(c('-o', '--outputpre'), default='CBF_',
help='output prefix (defaults to %default)'),
make_option(c('-a', '--antsCorticalThicknessPrefix'),
default='', help='prefix of antsCorticalThickness output'),
make_option(c('-l', '--labelSet'),
default='', help='label set in template space to warp to ASL'),
make_option(c('-t', '--template'),
default='', help='Template to warp output to'),
make_option(c('-c', '--paramFile'), default='',
help='parameter file containing ASL acquisition parameters'),
make_option(c('-x', '--smoothingFWHM'), default=0,
help='Full width half max for smoothing'),
make_option(c('-m', '--method'), default='regression',
help=paste(' method for perfusion calculation. \n\t\tOne of:',
'"regression", "subtraction", "bayesian",',
'"RobustRegression", "BayesianRegression", "LocalBayesianRegression."')),
make_option(c('-d', '--denoising'), default='CompCorMotion',
help=paste('denoising method.',
'Options are: "CompCor", "Motion", "Detrending",',
'\n\t\t"Cross-Validation", "OutlierRejection".',
'Multiple options can be specified',
'(e.g., "CompCorMotion" is legal). Default is %default.')),
make_option(c('-g', '--debug'), default=0,
help=paste('Save debugging information, including motion',
'correction and nuisance variables')),
make_option(c('-b', '--bloodT1'), default=0.67,
help='blood T1 value (defaults to %default s^-1)'),
make_option(c('-r', '--robustness'), default=0.95,
help='robustness parameter (defaults to %default).'),
make_option(c('-n', '--bootstrapNumber'), default=20,
help=' number of bootstrap samples (defaults to %default)'),
make_option(c('-e', '--bootstrapPercent'), default=0.70,
help='percent to sample per bootstrap run (defaults to %default)'),
make_option(c('-k', '--keepTmp'), default=0,
help=paste('keep tmp files, including warps',
'(defaults to %default--takes lots of space to save)')),
make_option(c('-f', '--bootstrapReplace'), default=0,
help=paste('bootstrap with replacement? takes arguments',
'0 or 1; defaults to 0.')),
make_option(c('-v', '--verbose'), default=0,
help='verbose output.'))
usage <- OptionParser(option_list=optlist, usage='Usage: %prog <s> [otlcxmdgbrnekfv]')
opt <- parse_args(usage)
## debug
#opt <- data.frame(
# pCASL=paste('/data/jag/BD2K01/ASL_pipeline/data/AddictionCenter/ABART/imgs/',
# '../processed/ABART_Bac_106/ASL/ABART_Bac_106_pCASL.nii.gz', sep=''),
# outputpre=paste('/data/jag/BD2K01/ASL_pipeline/data/AddictionCenter/ABART/imgs',
# '/../processed/ABART_Bac_106/ASL/ABART_Bac_106_', sep=''),
# antsCorticalThicknessPrefix=paste('/data/jag/BD2K01/ASL_pipeline/',
# 'data/AddictionCenter/ABART/imgs/../processed/ABART_Bac_106',
# '/ASL/../Anatomy/ABART_Bac_106_', sep=''),
# labelSet=paste('/data/jag/BD2K01/ASL_pipeline/templates/',
# 'HarvardOxford/ABART_rois.nii.gz', sep=''),
# template=paste('/data/jag/BD2K01/ASL_pipeline/templates/',
# 'HarvardOxford/MNI152_T1_2mm.nii.gz', sep=''))
# pCASL='data/101_pcasl.nii.gz',
# out='test')
if(!file.exists(as.character(opt$pCASL))) {
stop(paste('pCASL image', opt$pCASL,
'does not exist.'))
}
if(opt$verbose) {
cat('Running antsASLProcessing.R with the following options:\n')
for(option in names(opt)){
cat(paste(option, ': ', opt[option], '\n', sep=''))
}
}
if(length(grep(.Platform$file.sep, opt$outputpre)) > 0) {
outdir <- dirname(opt$outputpre)
if(!file.exists(outdir)) dir.create(outdir)
}
pcasl <- tryCatch({
antsImageRead(as.character(opt$pCASL), 4)
}, error = function(e) {
stop(paste('pCASL image', as.character(opt$pCASL),
'does not exist.'))
})
if(length(opt$paramFile) > 0){
if(file.exists(as.character(opt$paramFile))) {
config <- read.csv(opt$paramFile)
} else {
config <- data.frame(tagFirst=T, sequence='pcasl')
}
}
if (opt$smoothingFWHM > 0) {
mysmoother <- c(rep(opt$smoothingFWHM, 3), 0)
pcasl <- smoothImage(pcasl, mysmoother, FWHM=TRUE)
}
avg <- getAverageOfTimeSeries(pcasl)
avg <- n3BiasFieldCorrection(avg, 2)
avg <- n3BiasFieldCorrection(avg, 2)
mask <- getMask(avg, mean(avg), Inf, 2)
avg[mask==0] <- 0
moco <- antsrMotionCalculation(pcasl, fixed=avg, mask=mask)
tag.first <- config$tagFirst
ts <- timeseries2matrix(moco$moco_img, moco$moco_mask)
if (!tag.first) {
tc <- (rep(c(1, 0), dim(ts)[1])[1:dim(ts)[1]] - 0.5) # control minus tag
} else {
tc <- (rep(c(0, 1), dim(ts)[1])[1:dim(ts)[1]] - 0.5) # tag minus control
}
nuisance <- getASLNoisePredictors(ts, tc, polydegree='loess')
noise.all <- cbind(moco$moco_params, moco$fd$MeanDisplacement, nuisance)
noise.combined <- as.matrix(combineNuisancePredictors(ts, tc, noise.all))
onlypairs <- FALSE
if (opt$method == 'subtract') {
onlypairs <- TRUE
}
censored <- aslCensoring(pcasl, mask, nuis=noise.combined, method='robust',
reject.pairs=onlypairs)
if (length(censored$which.outliers) > 0) {
tc <- tc[-censored$which.outliers]
noise.censored <- noise.combined[-censored$which.outliers, ]
} else {
noise.censored <- noise.combined
}
if (opt$debug) {
mean.ts <- apply(ts, 1, mean)
dat.debug <- cbind(data.frame(MeanTimeSeries=mean.ts), noise.all)
write.csv(dat.debug, file=paste(opt$outputpre, 'TimeSeriesData.csv', sep=''),
row.names=as.character(1:nrow(ts)))
write.csv(data.frame(Outliers=censored$which.outliers),
file=paste(opt$outputpre, 'OutlierTimepoints.csv', sep=''))
}
if (opt$method == 'regression') {
perf <- aslAveraging(censored$asl.inlier, mask=moco$moco_mask,
tc=tc, nuisance=noise.censored, method='regression')
} else if (opt$method == 'bayesian') {
if (length(opt$antsCorticalThicknessPrefix) == 0) {
stop("For Bayesian regression, segmentations are required.")
}
act <- as.character(opt$antsCorticalThicknessPrefix)
braint1 <- tryCatch({
antsImageRead(paste(act, "ExtractedBrain0N4.nii.gz", sep=""))
}, error = function(e) {
print(paste('T1 brain image', paste(act, "ExtractedBrain0N4.nii.gz", sep=""),
'does not exist.'))
})
segmentation <- tryCatch({
antsImageRead(paste(act, "BrainSegmentation.nii.gz", sep=""))
}, error = function(e) {
stop(paste('Segmentation image', paste(act, "BrainSegmentation.nii.gz", sep=""),
'does not exist.'))
})
postnames <- list.files(path=dirname(act),
glob2rx("*BrainSegmentationPosteriors*.nii.gz"), full.names=TRUE)
tissuelist <- tryCatch({
imageFileNames2ImageList(postnames)
}, error = function(e) {
stop(paste("Probability images", postnames, "cannot be loaded."))
})
reg.t12asl <- antsRegistration(fixed=avg, moving=braint1,
typeofTransform="SyNBold", outprefix=as.character(opt$outputpre))
seg.asl <- antsApplyTransforms(avg, segmentation, reg.t12asl$fwdtransforms,
"MultiLabel")
for (ii in 1:length(tissuelist)) {
tissuelist[[ii]] <- antsApplyTransforms(avg, tissuelist[[ii]],
reg.t12asl$fwdtransforms, "Linear")
}
perf <- aslAveraging(censored$asl.inlier, mask=moco$moco_mask,
tc=tc, nuisance=noise.censored, method='bayesian',
segmentation=seg.asl, tissuelist=tissuelist)
} else if(opt$method == 'subtract'){
perf <- aslAveraging(censored$asl.inlier, mask=moco$moco_mask,
tc=tc, method='cubicSubtract')
}
mvals2 <- apply(ts[tc == 0.5, ], 2, mean)
mvals1 <- apply(ts[tc == -0.5, ], 2, mean)
# mean control should exceed mean tag
if (mean(mvals2) > mean(mvals1)) {
m0vals<-mvals2
m1vals<-mvals1
} else {
m0vals<-mvals1
m1vals<-mvals2
}
m0 <- antsImageClone(moco$moco_mask)
m0[moco$moco_mask == 0] <- 0
m0[moco$moco_mask == 1] <- m0vals
m0 <- n3BiasFieldCorrection(m0,4)
m0 <- n3BiasFieldCorrection(m0,2)
if (length(opt$config > 0)) {
tryCatch({
config <- read.csv(opt$config, row.names=1)
}, error = function(e){
print(paste("Configuration file", opt$config, "does not exist."))
})
parameters <- c(list(m0=antsImageClone(m0)), config)
} else {
parameters = list(sequence="pcasl", m0=antsImageClone(m0))
}
if (opt$debug) {
antsImageWrite(perf, paste(opt$outputpre, 'Perfusion.nii.gz', sep=''))
antsImageWrite(m0, paste(opt$outputpre, 'M0.nii.gz', sep=''))
}
cbf <- quantifyCBF(perf, mask=moco$moco_mask, parameters=parameters)
antsImageWrite(cbf$meancbf, paste(opt$outputpre, "CBF.nii.gz", sep=""))
if (nchar(opt$antsCorticalThicknessPrefix) > 0){
act <- as.character(opt$antsCorticalThicknessPrefix)
braint1 <- tryCatch({
antsImageRead(paste(act, "ExtractedBrain0N4.nii.gz", sep=""))
}, error = function(e) {
print(paste('T1 brain image', paste(act, "ExtractedBrain0N4.nii.gz", sep=""),
'does not exist.'))
})
seg <- tryCatch({
antsImageRead(paste(act, "BrainSegmentation.nii.gz", sep=""))
}, error = function(e) {
print(paste('Segmentation image', paste(act, "BrainSegmentation.nii.gz", sep=""),
'does not exist.'))
})
reg.t12asl <- antsRegistration(fixed=avg, moving=braint1,
typeofTransform="SyNBold" )
seg.asl <- antsApplyTransforms(avg, seg, reg.t12asl$fwdtransforms, "MultiLabel")
antsImageWrite(seg.asl, paste(opt$outputpre,
"SegmentationWarpedToASL.nii.gz", sep=''))
segstats <- labelStats(cbf$meancbf, seg.asl)
write.csv(segstats, paste(opt$outputpre, 'TissueStats.csv', sep=''),
row.names=FALSE)
tx.template2t1 <- c(paste(act, "TemplateToSubject0Warp.nii.gz", sep=""),
paste(act, "TemplateToSubject1GenericAffine.mat", sep=""))
tx.t12template <- c(paste(act, "SubjectToTemplate1Warp.nii.gz", sep=""),
paste(act, "SubjectToTemplate0GenericAffine.mat", sep=""))
tx.asl2template <- c(reg.t12asl$invtransforms, tx.t12template)
if (length(opt$template) > 0) {
template <- tryCatch({
antsImageRead(as.character(opt$template))
}, error = function(e) {
print(paste("Template image", template, "does not exist."))
})
asl.warped2template <- antsApplyTransforms(template, cbf$meancbf, tx.asl2template,
whichtoinvert=c(F, F, F, F))
antsImageWrite(asl.warped2template,
paste(opt$outputpre, "CBFWarpedToTemplate.nii.gz", sep=''))
}
tx.template2asl <- c(tx.template2t1, reg.t12asl$fwdtransforms)
if (nchar(as.character(opt$labelSet)) > 0) {
label <- tryCatch({
antsImageRead(as.character(opt$labelSet))
}, error = function(e) {
print(paste("Label image", opt$labelSet, "does not exist."))
})
label.asl <- antsApplyTransforms(avg, label, tx.template2asl, "MultiLabel")
antsImageWrite(label.asl, paste(opt$outputpre,
'LabelWarpedToASL.nii.gz', sep=''))
labelstats.cbf <- labelStats(cbf$meancbf, label.asl)
write.csv(labelstats.cbf, paste(opt$outputpre, 'LabelStats.csv', sep=''),
row.names=FALSE)
}
}