-
Notifications
You must be signed in to change notification settings - Fork 1
/
CCSegPipeSegToNative.py
executable file
·204 lines (154 loc) · 8.06 KB
/
CCSegPipeSegToNative.py
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
#!/usr/bin/env python3
import tempfile
import os
import scipy
import cv2
import math
import numpy
import nibabel
import matplotlib.pyplot as plt
import h5py
import shutil
import CCSegUtils
import CCSegPipeMidSagSymmetric
import subprocess
def segCCToNativeOneFile(segIMG, outputBase, midSagMATFile, segMATFile, dilateParaSagSlices = 0, flirtInterpType = 'trilinear'):
FID = h5py.File(segMATFile, 'r')
resampledAVWShape = numpy.array(FID['resampledAVWShape'])
LKCropRows = numpy.array(FID['LKCropRows'])
LKCropCols = numpy.array(FID['LKCropCols'])
midSagAVWxx = numpy.array(FID['midSagAVWxx'])
midSagAVWyy = numpy.array(FID['midSagAVWyy'])
resamplexx = numpy.array(FID['resamplexx'])
resampleyy = numpy.array(FID['resampleyy'])
#print seg['templatePixdims']
FID.close()
#print seg['IMG'].dtype
FID = h5py.File(midSagMATFile, 'r')
MSPMethod = str(numpy.array(FID['MSPMethod']))
if not MSPMethod == 'symmetric':
raise
midSagAVW = numpy.array(FID['midSagAVW'])
skullCrop = numpy.array(FID["skullCrop"]) # the initial cropping indices of the background
NIIOrnt = numpy.int16(numpy.array(FID["NIIOrnt"]))
NIIAffine = numpy.array(FID["NIIAffine"])
NIIShape = numpy.int16(numpy.array(FID["NIIShape"]))
axialNIIShape = numpy.int16(numpy.array(FID["axialNIIShape"]))
#axialNIIAffine = numpy.array(FID["axialNIIAffine"])
axialCroppedNIIShape = numpy.int16(numpy.array(FID["axialCroppedNIIShape"]))
midSlice = numpy.int16(numpy.array(FID["midSlice"]))
IMGxx = numpy.array(FID["IMGxx"])
IMGyy = numpy.array(FID["IMGyy"])
IMGzz = numpy.array(FID["IMGzz"])
finalRotY = numpy.array(FID["finalRotY"])
finalRotZ = numpy.array(FID["finalRotZ"])
finalTransX = numpy.array(FID["finalTransX"])
#midSagAVW = midSagAVW[:, ::-1]
#midSagAVW = numpy.rot90(midSagAVW, -1)
FID.close()
#plt.subplot(1, 4, 1); CCSegUtils.showIMG(seg['initialSeg'])
#plt.subplot(1, 4, 2); CCSegUtils.showIMG(seg['finalSeg'])
#plt.subplot(1, 4, 3); CCSegUtils.showIMG(seg['finalSegNoArtefacts'])
#plt.subplot(1, 4, 4); CCSegUtils.showIMG(seg['IMG'])
# this is going to be the segmentation in original resampled shape
finalSegResampledAVWSpace = numpy.zeros(resampledAVWShape, dtype = numpy.uint8)
# place the segmentation in the bounding box specified by the Lucas Kanade bounding box
finalSegResampledAVWSpace[LKCropRows[0]:(LKCropRows[1] + 1), LKCropCols[0]:(LKCropCols[1] + 1)] = segIMG
midSagAVWX, midSagAVWY = numpy.meshgrid(midSagAVWxx, midSagAVWyy)
# resample the midsagittal slice in resampled space to the original space
finalSegMidSagAVWSpace = CCSegUtils.interp2q(resamplexx, resampleyy, numpy.double(finalSegResampledAVWSpace), midSagAVWX, midSagAVWY, interpmethod = 'nearest', extrapval = 0)
#print finalSegMidSagAVWSpace.shape
#print axialCroppedNIIShape
#print axialNIIShape
#print skullCrop
#print axialCroppedNIIShape
finalSegNativeAxialCroppedSpace = numpy.zeros(axialCroppedNIIShape, dtype = numpy.int8)
#print axialNIIShape
#print finalSegNativeAxialCroppedSpace.shape
#print finalSegMidSagAVWSpace.shape
#quit()
finalSegNativeAxialCroppedSpace[midSlice] = numpy.rot90(finalSegMidSagAVWSpace, -1)
if dilateParaSagSlices > 0:
for z in range(dilateParaSagSlices):
finalSegNativeAxialCroppedSpace[numpy.maximum(midSlice - z - 1, 0)] = numpy.rot90(finalSegMidSagAVWSpace, -1)
finalSegNativeAxialCroppedSpace[numpy.minimum(midSlice + z + 1, finalSegNativeAxialCroppedSpace.shape[0])] = numpy.rot90(finalSegMidSagAVWSpace, -1)
#NIISaving = nibabel.Nifti1Image(numpy.uint8(finalSegNativeAxialCroppedSpace), axialNIIAffine)
#nibabel.save(NIISaving, outputBase + "_native_cropped_beforetransform.nii.gz")
#NIISaving = nibabel.Nifti1Image(finalSegNativeAxialCroppedSpace, NIIAffine)
#nibabel.save(NIISaving, outputBase + "_native_test_axial.nii.gz")
#print skullCrop
#print finalSegNativeAxialCroppedSpace.shape
#print IMGxx.shape
#print IMGyy.shape
#print IMGzz.shape
finalSegNativeAxialCroppedSpace = CCSegPipeMidSagSymmetric.transformIMG(finalSegNativeAxialCroppedSpace, IMGxx, IMGyy, IMGzz, -finalRotY, -finalRotZ, -finalTransX, interpmethod = 'nearest')
#NIISaving = nibabel.Nifti1Image(numpy.uint8(finalSegNativeAxialCroppedSpace), axialNIIAffine)
#nibabel.save(NIISaving, outputBase + "_native_cropped_aftertransform.nii.gz")
#finalSegNativeAxialCroppedSpace = numpy.rot90(finalSegNativeAxialCroppedSpace, 1)
#print skullCrop
#print finalSegNativeAxialCroppedSpace.shape
#print axialNIIShape
#print NIIShape
# uncrop using skull crop
#finalSegNativeAxialSpace = numpy.zeros((NIIShape[1], NIIShape[0], NIIShape[2]), dtype = numpy.int8)
finalSegNativeAxialSpace = numpy.zeros(axialNIIShape, dtype = numpy.int8)
finalSegNativeAxialSpace[skullCrop[0, 0]:(skullCrop[0, 1] + 1), skullCrop[1, 0]:(skullCrop[1, 1] + 1), skullCrop[2, 0]:(skullCrop[2, 1] + 1)] = numpy.array(finalSegNativeAxialCroppedSpace)
#finalSegNativeAxialSpace = numpy.rot90(finalSegNativeAxialSpace, -1);
# transform to original orientation
MNITemplate = CCSegUtils.MNI152FLIRTTemplate(skullStripped = False)
MNITemplateNII = nibabel.load(MNITemplate)
MNIAXCodes = nibabel.aff2axcodes(MNITemplateNII.affine)
MNIOrnt = nibabel.orientations.axcodes2ornt(MNIAXCodes)
MNIToNIITransformOrnt = nibabel.orientations.ornt_transform(MNIOrnt, NIIOrnt)
nativeNIIIMG = nibabel.orientations.apply_orientation(finalSegNativeAxialSpace, MNIToNIITransformOrnt)
NIISaving = nibabel.Nifti1Image(nativeNIIIMG, NIIAffine)
nibabel.save(NIISaving, outputBase + "_native.nii.gz")
def segCCToNative(outputBase, doGraphics = False, dilateParaSagSlices = 0):
midSagMATFile = outputBase + "_midsag.hdf5"
assert(os.path.isfile(midSagMATFile)),"midsag hdf5 file not found"
segMATFile = outputBase + "_seg.hdf5"
segManeditMATFile = outputBase + "_seg_manedit.hdf5"
#finalSeg = numpy.array(seg['finalSeg'])
assert(os.path.isfile(segMATFile)),"Seg mat file not found: " + segMATFile
FID = h5py.File(segMATFile, 'r')
autoFinalSeg = numpy.array(FID['finalSegNoArtefacts'])
templatePixdims = numpy.array(FID['templatePixdims'])
#print seg['templatePixdims']
FID.close()
#print seg['IMG'].dtype
#plt.subplot(1, 4, 1); CCSegUtils.showIMG(seg['initialSeg'])
#plt.subplot(1, 4, 2); CCSegUtils.showIMG(seg['finalSeg'])
#plt.subplot(1, 4, 3); CCSegUtils.showIMG(seg['finalSegNoArtefacts'])
#plt.subplot(1, 4, 4); CCSegUtils.showIMG(seg['IMG'])
#plt.gcf().set_size_inches((20, 10), forward = True)
#plt.show()
#quit()
if os.path.isfile(segManeditMATFile):
FID = h5py.File(segManeditMATFile, 'r')
finalSeg = numpy.array(FID['finalSegManEdit']) > 0
FID.close()
else:
finalSeg = numpy.array(autoFinalSeg)
segCCToNativeOneFile(numpy.uint8(finalSeg), outputBase + "_seg", midSagMATFile, segMATFile, dilateParaSagSlices = dilateParaSagSlices)
thicknessMATFile = outputBase + "_thickness.hdf5"
if os.path.isfile(thicknessMATFile):
finalSegxx = numpy.arange(finalSeg.shape[1])
finalSegyy = numpy.arange(finalSeg.shape[0])
finalSegX, finalSegY = numpy.meshgrid(finalSegxx, finalSegyy)
FID = h5py.File(thicknessMATFile, 'r')
labelNames = ['witelson', 'hoferFrahm', 'emsell']
for curLabels in labelNames:
curLabelIMG = numpy.array(FID[curLabels + 'Labels'])
# for the parcellations, we need to get the parcellation images in the same space as the segmentation images
# so we resample the parcellation image using the grid of the segmentation image, we need to reverse a scaling that was performed in the thickness part:
curLabelIMG = CCSegUtils.interp2q(numpy.array(FID['xx']) / templatePixdims[0], numpy.array(FID['yy']) / templatePixdims[1], numpy.double(curLabelIMG), finalSegX, finalSegY, interpmethod = 'nearest', extrapval = 0)
#SR = 1
#SC = 3
#plt.subplot(SR, SC, 1); CCSegUtils.showIMG(finalSeg)
#plt.subplot(SR, SC, 2); CCSegUtils.showIMG(curLabelIMG)
#plt.subplot(SR, SC, 3); CCSegUtils.showIMG(T)
#plt.gcf().set_size_inches((20, 10), forward = True)
#plt.show()
#quit()
segCCToNativeOneFile(numpy.uint8(curLabelIMG), outputBase + "_parcellation_" + curLabels.lower(), midSagMATFile, segMATFile, flirtInterpType = 'nearestneighbour', dilateParaSagSlices = dilateParaSagSlices)
FID.close()