-
Notifications
You must be signed in to change notification settings - Fork 7
/
neonate-preprocessing-v2.sh
executable file
·179 lines (134 loc) · 7.83 KB
/
neonate-preprocessing-v2.sh
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
#!/bin/bash
#Preprocessing script for mouse brains, using Dorr-steadman template and MINC files converted from DICOM from Paravision 5.x
#If using conversion from PV6, might need to remove the volflip -y command
set -euo pipefail
set -x
calc(){ awk "BEGIN { print "$*" }"; }
tmpdir=$(mktemp -d)
input=$1
output=$2
origdistance=20
distance=${origdistance}
levels=4
cycles=3
iters=25
lambda=2e-6
shrink=4.0
fwhm=0.1
stop=1e-5
do_correct() {
distance=${origdistance}
j=0
while (( j < levels )); do
i=0
while (( i < cycles )); do
nu_correct -clobber -normalize_field \
-stop ${stop} -distance ${distance} -iterations ${iters} -fwhm ${fwhm} -shrink ${shrink} -lambda ${lambda} \
-mask ${tmpdir}/weight.mnc ${n3input} ${tmpdir}/corrected_${distance}_${i}.mnc
n3input=${tmpdir}/corrected_${distance}_${i}.mnc
((++i))
done
distance=$(calc ${distance}/2)
((++j))
done
}
#modelfile=${QUARANTINE_PATH}/resources/in-vivo-MEMRI_90micron/Dorr_2008_on_MEMRI_C57BL6_P43_average_recenter.mnc
#modelmask=${QUARANTINE_PATH}/resources/in-vivo-MEMRI_90micron/Dorr_2008_on_MEMRI_C57BL6_P43_mask_recenter.mnc
modelfile=${QUARANTINE_PATH}/resources/P7_ECox_2015/N4/P7_ECox.mnc
modelmask=${QUARANTINE_PATH}/resources/P7_ECox_2015/N4/P7_ECox_mask.mnc
cp -f ${input} ${tmpdir}/input.mnc
input=${tmpdir}/input.mnc
minimum_resolution=$(python -c "print(min([abs(x) for x in [float(x) for x in \"$(PrintHeader ${input} 1)\".split(\"x\")]]))")
minc_modify_header $input -sinsert :history=''
#Forceably convert to MINC2, and clamp range to avoid negative numbers, rescale to 0-65535
mincconvert -2 ${input} ${tmpdir}/originput.mnc
#Rescale initial data into entirely positive range (fix for completely negative data)
ImageMath 3 ${tmpdir}/originput.mnc RescaleImage ${tmpdir}/originput.mnc 0 65535
#Very mild range clamp for very hot voxels
mincmath -quiet ${N4_VERBOSE:+-verbose} -clamp \
-const2 $(mincstats -quiet -floor 1e-12 -pctT 0.01 ${tmpdir}/originput.mnc) \
$(mincstats -quiet -floor 1e-12 -pctT 99.99 ${tmpdir}/originput.mnc) \
${tmpdir}/originput.mnc ${tmpdir}/originput.clamp.mnc
ImageMath 3 ${tmpdir}/originput.clamp.mnc RescaleImage ${tmpdir}/originput.clamp.mnc 0 65535
mincresample -quiet ${N4_VERBOSE:+-verbose} -like ${tmpdir}/originput.mnc -keep -unsigned -short \
${tmpdir}/originput.clamp.mnc ${tmpdir}/originput.clamp.resample.mnc
mv -f ${tmpdir}/originput.clamp.resample.mnc ${tmpdir}/originput.mnc
rm -f ${tmpdir}/originput.clamp.mnc
#Reorient to MINC RAS
#volmash -swap zy ${tmpdir}/originput.mnc ${tmpdir}/mash.mnc
#volflip -x ${tmpdir}/mash.mnc ${tmpdir}/flip.mnc
#cp -f ${tmpdir}/flip.mnc ${tmpdir}/originput.mnc
#Construct Otsu Mask of entire image
input=${tmpdir}/originput.mnc
volcentre -zero_dircos ${input} ${tmpdir}/centre.mnc
mv -f ${tmpdir}/centre.mnc ${input}
ImageMath 3 ${tmpdir}/originput.mnc PadImage ${tmpdir}/originput.mnc 20
ThresholdImage 3 ${input} ${tmpdir}/bgmask.mnc 1e-12 Inf 1 0
ThresholdImage 3 ${input} ${tmpdir}/weight.mnc Otsu 4 ${tmpdir}/bgmask.mnc
ThresholdImage 3 ${tmpdir}/weight.mnc ${tmpdir}/weight.mnc 2 Inf 1 0
iMath 3 ${tmpdir}/weight.mnc ME ${tmpdir}/weight.mnc 1 1 ball 1
ImageMath 3 ${tmpdir}/weight.mnc GetLargestComponent ${tmpdir}/weight.mnc
iMath 3 ${tmpdir}/weight.mnc MD ${tmpdir}/weight.mnc 1 1 ball 1
ExtractRegionFromImageByMask 3 ${input} ${tmpdir}/repad.mnc ${tmpdir}/weight.mnc 1 10
ExtractRegionFromImageByMask 3 ${tmpdir}/originput.mnc ${tmpdir}/repad_orig.mnc ${tmpdir}/weight.mnc 1 10
mv -f ${tmpdir}/repad_orig.mnc ${tmpdir}/originput.mnc
mv -f ${tmpdir}/repad.mnc ${input}
ThresholdImage 3 ${input} ${tmpdir}/bgmask.mnc 1e-12 Inf 1 0
mincresample -like ${input} -keep -near -labels ${tmpdir}/weight.mnc ${tmpdir}/weight2.mnc
mv -f ${tmpdir}/weight2.mnc ${tmpdir}/weight.mnc
minc_anlm --clobber --mt $(nproc) ${input} ${tmpdir}/denoise.mnc
n3input=${tmpdir}/denoise.mnc
do_correct
ThresholdImage 3 ${n3input} ${tmpdir}/weight.mnc Otsu 4 ${tmpdir}/bgmask.mnc
ThresholdImage 3 ${tmpdir}/weight.mnc ${tmpdir}/weight.mnc 2 Inf 1 0
iMath 3 ${tmpdir}/weight.mnc ME ${tmpdir}/weight.mnc 1 1 ball 1
ImageMath 3 ${tmpdir}/weight.mnc GetLargestComponent ${tmpdir}/weight.mnc
iMath 3 ${tmpdir}/weight.mnc MD ${tmpdir}/weight.mnc 1 1 ball 1
mincresample -like ${input} -keep -near -labels ${tmpdir}/weight.mnc ${tmpdir}/weight2.mnc
mv -f ${tmpdir}/weight2.mnc ${tmpdir}/weight.mnc
n3input=${tmpdir}/denoise.mnc
do_correct
for file in ${tmpdir}/*imp; do
echo nu_evaluate -clobber -mapping ${file} -mask ${tmpdir}/weight.mnc -field ${tmpdir}/$(basename $file .imp)_field.mnc ${input} ${tmpdir}/$(basename $file .imp).mnc
done | parallel
mincmath -clobber -mult ${tmpdir}/*field.mnc ${tmpdir}/precorrect_field_combined.mnc
mincmath -clobber -copy_header -zero -div ${tmpdir}/originput.mnc ${tmpdir}/precorrect_field_combined.mnc ${tmpdir}/precorrect.mnc
minc_anlm --clobber --mt $(nproc) ${tmpdir}/precorrect.mnc ${tmpdir}/denoise.mnc
n3input=${tmpdir}/denoise.mnc
antsRegistration_affine_SyN.sh --verbose --no-histogram-matching --fast \
${n3input} ${modelfile} ${tmpdir}/tomodel
antsRegistration_affine_SyN.sh --clobber --skip-linear --initial-transform ${tmpdir}/tomodel0_GenericAffine.xfm \
--verbose --no-histogram-matching --fast --fixed-mask ${modelmask} \
${n3input} ${modelfile} ${tmpdir}/tomodel
antsApplyTransforms -d 3 -i ${modelmask} -t [ ${tmpdir}/tomodel0_GenericAffine.xfm,1 ] -t ${tmpdir}/tomodel1_inverse_NL.xfm \
-o ${tmpdir}/newmask.mnc -n GenericLabel -r ${n3input} --verbose
iMath 3 ${tmpdir}/newmask.mnc MD ${tmpdir}/newmask.mnc 1 1 ball 1
ThresholdImage 3 ${n3input} ${tmpdir}/weight.mnc Otsu 4 ${tmpdir}/newmask.mnc
ThresholdImage 3 ${tmpdir}/weight.mnc ${tmpdir}/weight.mnc 2 Inf 1 0
ImageMath 3 ${tmpdir}/weight.mnc m ${tmpdir}/newmask.mnc ${tmpdir}/weight.mnc
iMath 3 ${tmpdir}/weight.mnc ME ${tmpdir}/weight.mnc 1 1 ball 1
ImageMath 3 ${tmpdir}/weight.mnc GetLargestComponent ${tmpdir}/weight.mnc
iMath 3 ${tmpdir}/weight.mnc MD ${tmpdir}/weight.mnc 1 1 ball 1
mincresample -like ${input} -keep -near -labels ${tmpdir}/weight.mnc ${tmpdir}/weight2.mnc
mv -f ${tmpdir}/weight2.mnc ${tmpdir}/weight.mnc
do_correct
for file in ${tmpdir}/*imp; do
echo nu_evaluate -clobber -mapping ${file} -mask ${tmpdir}/weight.mnc -field ${tmpdir}/$(basename $file .imp)_field.mnc ${input} ${tmpdir}/$(basename $file .imp).mnc
done | parallel
mincmath -clobber -mult ${tmpdir}/*field.mnc ${tmpdir}/precorrect_field_combined.mnc ${tmpdir}/field_final.mnc
mincmath -clobber -copy_header -zero -div ${tmpdir}/originput.mnc ${tmpdir}/field_final.mnc ${tmpdir}/correct.mnc
minc_anlm --mt $(nproc) ${tmpdir}/correct.mnc ${tmpdir}/denoise_correct.mnc
ExtractRegionFromImageByMask 3 ${tmpdir}/denoise_correct.mnc ${tmpdir}/recrop.mnc ${tmpdir}/weight.mnc 1 10
mv -f ${tmpdir}/recrop.mnc ${tmpdir}/denoise_correct.mnc
mincresample -keep -near -like ${tmpdir}/denoise_correct.mnc ${tmpdir}/weight.mnc ${tmpdir}/weight2.mnc
mv -f ${tmpdir}/weight2.mnc ${tmpdir}/weight.mnc
cp ${tmpdir}/weight.mnc $(dirname ${output})/$(basename ${output} .mnc)_mask.mnc
cp ${tmpdir}/denoise_correct.mnc ${output}
xfminvert ${tmpdir}/tomodel0_GenericAffine.xfm ${tmpdir}/tomodel0_GenericAffine_invert.xfm
param2xfm $(xfm2param ${tmpdir}/tomodel0_GenericAffine_invert.xfm | grep -E 'scale|shear') ${tmpdir}/scaleshear.xfm
xfminvert ${tmpdir}/scaleshear.xfm ${tmpdir}/unscaleshear.xfm
xfmconcat ${tmpdir}/tomodel0_GenericAffine_invert.xfm ${tmpdir}/unscaleshear.xfm ${tmpdir}/lsq6.xfm
mincresample -tfm_input_sampling -transform ${tmpdir}/lsq6.xfm ${tmpdir}/denoise_correct.mnc ${tmpdir}/lsq6.mnc
mincmath -clamp -const2 0 $(mincstats -quiet -max ${tmpdir}/lsq6.mnc) ${tmpdir}/lsq6.mnc $(dirname ${output})/$(basename ${output} .mnc)_lsq6.mnc
mincresample -transform ${tmpdir}/lsq6.xfm -like $(dirname ${output})/$(basename ${output} .mnc)_lsq6.mnc -keep -near -labels ${tmpdir}/weight.mnc $(dirname ${output})/$(basename ${output} .mnc)_lsq6_mask.mnc
rm -rf ${tmpdir}