-
Notifications
You must be signed in to change notification settings - Fork 0
/
data preprocessing.py
250 lines (196 loc) · 9.67 KB
/
data preprocessing.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
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
from skimage.transform import resize
import numpy as np
import os
import SimpleITK as sitk
import h5py
def getFiles(targetdir):
ls = []
for fname in os.listdir(targetdir):
path = os.path.join(targetdir, fname)
if os.path.isdir(path):
continue
ls.append(fname)
return ls
def create_nonzero_mask(data):
from scipy.ndimage import binary_fill_holes
assert len(data.shape) == 4 or len(data.shape) == 3, \
"data must have shape (C, X, Y, Z) or shape (C, X, Y)"
nonzero_mask = np.zeros(data.shape[1:], dtype=bool)
for c in range(data.shape[0]):
this_mask = data[c] >= -325
nonzero_mask = nonzero_mask | this_mask
nonzero_mask = binary_fill_holes(nonzero_mask)
return nonzero_mask
# 获取bounding box的函数。这个在日常会经常用到
def get_bbox_from_mask(mask, outside_value=0):
mask_voxel_coords = np.where(mask != outside_value)
minzidx = int(np.min(mask_voxel_coords[0]))
maxzidx = int(np.max(mask_voxel_coords[0])) + 1
minxidx = int(np.min(mask_voxel_coords[1]))
maxxidx = int(np.max(mask_voxel_coords[1])) + 1
minyidx = int(np.min(mask_voxel_coords[2]))
maxyidx = int(np.max(mask_voxel_coords[2])) + 1
return [[minzidx, maxzidx], [minxidx, maxxidx], [minyidx, maxyidx]]
# 根据bbox提取ROI，这个也经常会用到
def crop_to_bbox(image, bbox):
assert len(image.shape) == 3, "only supports 3d images"
resizer = (slice(bbox[0][0], bbox[0][1]), slice(bbox[1][0], bbox[1][1]), slice(bbox[2][0], bbox[2][1]))
return image[resizer]
def _compute_stats(voxels):
if len(voxels) == 0:
return np.nan, np.nan, np.nan, np.nan, np.nan, np.nan, np.nan
median = np.median(voxels)
mean = np.mean(voxels)
sd = np.std(voxels)
mn = np.min(voxels)
mx = np.max(voxels)
percentile_99_5 = np.percentile(voxels, 99.5)
percentile_00_5 = np.percentile(voxels, 00.5)
return median, mean, sd, mn, mx, percentile_99_5, percentile_00_5
def resampleVolume(outspacing, vol):
"""
将体数据重采样的指定的spacing大小\n
paras：
outpacing：指定的spacing，例如[1,1,1]
vol：sitk读取的image信息，这里是体数据\n
return：重采样后的数据
"""
outsize = [0, 0, 0]
# 读取文件的size和spacing信息
inputsize = vol.GetSize()
inputspacing = vol.GetSpacing()
# 计算改变spacing后的size，用物理尺寸/体素的大小
outsize[0] = round(inputsize[0] * inputspacing[0] / outspacing[0])
outsize[1] = round(inputsize[1] * inputspacing[1] / outspacing[1])
outsize[2] = round(inputsize[2] * inputspacing[2] / outspacing[2])
outsize = [outsize[2], outsize[1], outsize[0]]
# 设定重采样的一些参数
image_data = sitk.GetArrayFromImage(vol)
im = resize(image_data, output_shape=outsize, order=3, mode='edge', anti_aliasing=False)
image = sitk.GetImageFromArray(im)
image.SetSpacing(outspacing)
image.SetDirection(vol.GetDirection())
image.SetOrigin(vol.GetOrigin())
return image
def resampleMask(outspacing, vol):
"""
将体数据重采样的指定的spacing大小\n
paras：
outpacing：指定的spacing，例如[1,1,1]
vol：sitk读取的image信息，这里是体数据\n
return：重采样后的数据
"""
outsize = [0, 0, 0]
# 读取文件的size和spacing信息
inputsize = vol.GetSize()
inputspacing = vol.GetSpacing()
# 计算改变spacing后的size，用物理尺寸/体素的大小
outsize[0] = round(inputsize[0] * inputspacing[0] / outspacing[0])
outsize[1] = round(inputsize[1] * inputspacing[1] / outspacing[1])
outsize[2] = round(inputsize[2] * inputspacing[2] / outspacing[2])
outsize = [outsize[2], outsize[1], outsize[0]]
# 设定重采样的一些参数
image_data = sitk.GetArrayFromImage(vol)
im = resize(image_data, output_shape=outsize, order=0, anti_aliasing=False)
image = sitk.GetImageFromArray(im)
image.SetSpacing(outspacing)
image.SetDirection(vol.GetDirection())
image.SetOrigin(vol.GetOrigin())
return image
def normalize_data(data): # flare
data = np.clip(data, -22.0, 325.0)
data = np.array(data, dtype=np.float32)
data -= 214.68231
data /= 100.240135
return data
def resample_label_image(label_image, new_spacing, interpolator=sitk.sitkNearestNeighbor):
original_spacing = label_image.GetSpacing()
original_size = label_image.GetSize()
new_size = [int(round(osz * ospc / nspc)) for osz, ospc, nspc in zip(original_size, original_spacing, new_spacing)]
resample = sitk.ResampleImageFilter()
resample.SetOutputSpacing(new_spacing)
resample.SetSize(new_size)
resample.SetOutputDirection(label_image.GetDirection())
resample.SetOutputOrigin(label_image.GetOrigin())
resample.SetInterpolator(interpolator)
return resample.Execute(label_image)
imdir = "F:/MMWHS_evaluation_testdata_label_encrypt_1mm_forpublic/imagesTr/" # DIRECTORY PATH TO data
labeldir = "F:/MMWHS_evaluation_testdata_label_encrypt_1mm_forpublic/labelsTr/" # DIRECTORY PATH TO label
outdir = "F:/MMWHS_evaluation_testdata_label_encrypt_1mm_forpublic/h555/imageout/" # DIRECTORY PATH TO processed data nii
outdir2 = "F:/MMWHS_evaluation_testdata_label_encrypt_1mm_forpublic/h555/labelout//" # DIRECTORY PATH TO processed label nii
out_dir = "F:/MMWHS_evaluation_testdata_label_encrypt_1mm_forpublic/h555/h5/" # DIRECTORY PATH TO processed h5
for pdx, fname in enumerate(sorted(getFiles(imdir))):
sitk_im = sitk.ReadImage(os.path.join(imdir, fname))
fname = fname.replace('_0000', '')
sitk_mask = sitk.ReadImage(os.path.join(labeldir, fname))
img_fdata = sitk.GetArrayFromImage(sitk_im)
img_fdata = img_fdata.astype(np.float32)
mask_fdata = sitk.GetArrayFromImage(sitk_mask)
directions = np.asarray(sitk_im.GetDirection())
directions = directions.tolist()
if directions == [1, 0, 0, 0, -1, 0, 0, 0, -1]:
img_fdata = np.flip(img_fdata, [0, 2])
mask_fdata = np.flip(mask_fdata, [0, 2])
data = np.clip(img_fdata, -325, 325)
saveimg = sitk.GetImageFromArray(data)
saveimg.SetSpacing(sitk_im.GetSpacing())
saveimg.SetDirection([-1, 0, 0, 0, -1, 0, 0, 0, 1])
saveimg.SetOrigin(sitk_im.GetOrigin())
vol_resampled = resampleVolume([1.22250766, 1.22250766, 2.5], saveimg) #flare
savemask = sitk.GetImageFromArray(mask_fdata)
savemask.SetSpacing(sitk_im.GetSpacing())
savemask.SetDirection([-1, 0, 0, 0, -1, 0, 0, 0, 1])
savemask.SetOrigin(sitk_im.GetOrigin())
vol_resampled_mask = resampleMask([1.22250766, 1.22250766, 2.5], savemask) #flare
resize_imgarr = sitk.GetArrayFromImage(vol_resampled) # 96 128 128
nor_resize_imgarr = normalize_data(resize_imgarr)
nor_resize_img = sitk.GetImageFromArray(nor_resize_imgarr)
nor_resize_img.SetSpacing(vol_resampled.GetSpacing())
nor_resize_img.SetOrigin(vol_resampled.GetOrigin())
nor_resize_img.SetDirection([-1, 0, 0, 0, -1, 0, 0, 0, 1])
print(nor_resize_img.GetDirection())
sitk.WriteImage(nor_resize_img, outdir + fname.split('.')[0] + '.nii.gz')
sitk.WriteImage(vol_resampled_mask, outdir2 + fname.split('.')[0] + '.nii.gz')
# to h5
savename = out_dir + fname.split('.')[0]
if not os.path.isdir(savename):
os.makedirs(savename)
im = sitk.GetArrayFromImage(nor_resize_img)
label = sitk.GetArrayFromImage(vol_resampled_mask)
directions = np.asarray(nor_resize_img.GetDirection())
f = h5py.File(savename + '/2022.h5', 'w')
f.create_dataset('image', data=im, compression="gzip")
f.create_dataset('label', data=label, compression="gzip")
f.close()
else:
data = np.clip(img_fdata, -325, 325)
saveimg = sitk.GetImageFromArray(img_fdata)
saveimg.SetSpacing(sitk_im.GetSpacing())
saveimg.SetDirection(sitk_im.GetDirection())
saveimg.SetOrigin(sitk_im.GetOrigin())
vol_resampled = resampleVolume([1.22250766, 1.22250766, 2.5], saveimg) #flare
savemask = sitk.GetImageFromArray(mask_fdata)
savemask.SetSpacing(sitk_im.GetSpacing())
savemask.SetDirection(sitk_im.GetDirection())
savemask.SetOrigin(sitk_im.GetOrigin())
vol_resampled_mask = resampleMask([1.22250766, 1.22250766, 2.5], savemask) #flare
resize_imgarr = sitk.GetArrayFromImage(vol_resampled) # 96 128 128
nor_resize_imgarr = normalize_data(resize_imgarr)
nor_resize_img = sitk.GetImageFromArray(nor_resize_imgarr)
nor_resize_img.SetSpacing(vol_resampled.GetSpacing())
nor_resize_img.SetOrigin(vol_resampled.GetOrigin())
nor_resize_img.SetDirection(vol_resampled.GetDirection())
print(nor_resize_img.GetDirection())
sitk.WriteImage(nor_resize_img, outdir + fname.split('.')[0] + '.nii.gz')
sitk.WriteImage(vol_resampled_mask, outdir2 + fname.split('.')[0] + '.nii.gz')
# to h5
savename = out_dir + fname.split('.')[0]
if not os.path.isdir(savename):
os.makedirs(savename)
im = sitk.GetArrayFromImage(nor_resize_img)
label = sitk.GetArrayFromImage(vol_resampled_mask)
directions = np.asarray(nor_resize_img.GetDirection())
f = h5py.File(savename + '/2022.h5', 'w')
f.create_dataset('image', data=im, compression="gzip")
f.create_dataset('label', data=label, compression="gzip")
f.close()