project-monai commited on
Commit
0798b5a
·
verified ·
1 Parent(s): abec4fb

Upload multi_organ_segmentation version 0.0.5

Browse files
.gitattributes CHANGED
@@ -33,3 +33,5 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ docs/output_example.png filter=lfs diff=lfs merge=lfs -text
37
+ models/model.ts filter=lfs diff=lfs merge=lfs -text
LICENSE ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ Private Use, Non-Commercial, Non-Reverse-Engineering License
2
+
3
+ The Licensee is not allowed to distribute or make the model to any third party, either for free or for a fee. Reverse engineering of the model is not allowed. This includes, but is not limited to, providing the model as part of a commercial offering, sharing the model on a public or private network, or making the model available for download on the Internet.
configs/dataset_0.json ADDED
@@ -0,0 +1,1688 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "training": [
3
+ {
4
+ "label": "ACC_g559h/ACC_g559h-label.nii.gz",
5
+ "image": "ACC_g559h/ACC_g559h-data.nii.gz"
6
+ },
7
+ {
8
+ "label": "ACC_g742h/ACC_g742h-label.nii.gz",
9
+ "image": "ACC_g742h/ACC_g742h-data.nii.gz"
10
+ },
11
+ {
12
+ "label": "ACC_g631h/ACC_g631h-label.nii.gz",
13
+ "image": "ACC_g631h/ACC_g631h-data.nii.gz"
14
+ },
15
+ {
16
+ "label": "ACC_g316h/ACC_g316h-label.nii.gz",
17
+ "image": "ACC_g316h/ACC_g316h-data.nii.gz"
18
+ },
19
+ {
20
+ "label": "ACC_g330h/ACC_g330h-label.nii.gz",
21
+ "image": "ACC_g330h/ACC_g330h-data.nii.gz"
22
+ },
23
+ {
24
+ "label": "ACC_g619h/ACC_g619h-label.nii.gz",
25
+ "image": "ACC_g619h/ACC_g619h-data.nii.gz"
26
+ },
27
+ {
28
+ "label": "ACC_g456h/ACC_g456h-label.nii.gz",
29
+ "image": "ACC_g456h/ACC_g456h-data.nii.gz"
30
+ },
31
+ {
32
+ "label": "acc_g051h/acc_g051h-label.nii.gz",
33
+ "image": "acc_g051h/acc_g051h-data.nii.gz"
34
+ },
35
+ {
36
+ "label": "ACC_g428h/ACC_g428h-label.nii.gz",
37
+ "image": "ACC_g428h/ACC_g428h-data.nii.gz"
38
+ },
39
+ {
40
+ "label": "ACC_g383h/ACC_g383h-label.nii.gz",
41
+ "image": "ACC_g383h/ACC_g383h-data.nii.gz"
42
+ },
43
+ {
44
+ "label": "ACC_g376h/ACC_g376h-label.nii.gz",
45
+ "image": "ACC_g376h/ACC_g376h-data.nii.gz"
46
+ },
47
+ {
48
+ "label": "ACC_g728h/ACC_g728h-label.nii.gz",
49
+ "image": "ACC_g728h/ACC_g728h-data.nii.gz"
50
+ },
51
+ {
52
+ "label": "ACC_g658h/ACC_g658h-label.nii.gz",
53
+ "image": "ACC_g658h/ACC_g658h-data.nii.gz"
54
+ },
55
+ {
56
+ "label": "ACC_g639h/ACC_g639h-label.nii.gz",
57
+ "image": "ACC_g639h/ACC_g639h-data.nii.gz"
58
+ },
59
+ {
60
+ "label": "ACC_g410h/ACC_g410h-label.nii.gz",
61
+ "image": "ACC_g410h/ACC_g410h-data.nii.gz"
62
+ },
63
+ {
64
+ "label": "ACC_g153h/ACC_g153h-label.nii.gz",
65
+ "image": "ACC_g153h/ACC_g153h-data.nii.gz"
66
+ },
67
+ {
68
+ "label": "ACC_g478h/ACC_g478h-label.nii.gz",
69
+ "image": "ACC_g478h/ACC_g478h-data.nii.gz"
70
+ },
71
+ {
72
+ "label": "ACC_g406h/ACC_g406h-label.nii.gz",
73
+ "image": "ACC_g406h/ACC_g406h-data.nii.gz"
74
+ },
75
+ {
76
+ "label": "ACC_g114h/ACC_g114h-label.nii.gz",
77
+ "image": "ACC_g114h/ACC_g114h-data.nii.gz"
78
+ },
79
+ {
80
+ "label": "ACC_g142h/ACC_g142h-label.nii.gz",
81
+ "image": "ACC_g142h/ACC_g142h-data.nii.gz"
82
+ },
83
+ {
84
+ "label": "acc_g031/acc_g031-label.nii.gz",
85
+ "image": "acc_g031/acc_g031-data.nii.gz"
86
+ },
87
+ {
88
+ "label": "ACC_g432h/ACC_g432h-label.nii.gz",
89
+ "image": "ACC_g432h/ACC_g432h-data.nii.gz"
90
+ },
91
+ {
92
+ "label": "ACC_g636h/ACC_g636h-label.nii.gz",
93
+ "image": "ACC_g636h/ACC_g636h-data.nii.gz"
94
+ },
95
+ {
96
+ "label": "ACC_g522h/ACC_g522h-label.nii.gz",
97
+ "image": "ACC_g522h/ACC_g522h-data.nii.gz"
98
+ },
99
+ {
100
+ "label": "ACC_g676h/ACC_g676h-label.nii.gz",
101
+ "image": "ACC_g676h/ACC_g676h-data.nii.gz"
102
+ },
103
+ {
104
+ "label": "acc_g075h/acc_g075h-label.nii.gz",
105
+ "image": "acc_g075h/acc_g075h-data.nii.gz"
106
+ },
107
+ {
108
+ "label": "ACC_g280h/ACC_g280h-label.nii.gz",
109
+ "image": "ACC_g280h/ACC_g280h-data.nii.gz"
110
+ },
111
+ {
112
+ "label": "ACC_g496h/ACC_g496h-label.nii.gz",
113
+ "image": "ACC_g496h/ACC_g496h-data.nii.gz"
114
+ },
115
+ {
116
+ "label": "ACC_g442h/ACC_g442h-label.nii.gz",
117
+ "image": "ACC_g442h/ACC_g442h-data.nii.gz"
118
+ },
119
+ {
120
+ "label": "ACC_g537h/ACC_g537h-label.nii.gz",
121
+ "image": "ACC_g537h/ACC_g537h-data.nii.gz"
122
+ },
123
+ {
124
+ "label": "ACC_g122h/ACC_g122h-label.nii.gz",
125
+ "image": "ACC_g122h/ACC_g122h-data.nii.gz"
126
+ },
127
+ {
128
+ "label": "ACC_g140h/ACC_g140h-label.nii.gz",
129
+ "image": "ACC_g140h/ACC_g140h-data.nii.gz"
130
+ },
131
+ {
132
+ "label": "ACC_g220h/ACC_g220h-label.nii.gz",
133
+ "image": "ACC_g220h/ACC_g220h-data.nii.gz"
134
+ },
135
+ {
136
+ "label": "acc_g039/acc_g039-label.nii.gz",
137
+ "image": "acc_g039/acc_g039-data.nii.gz"
138
+ },
139
+ {
140
+ "label": "ACC_g375h/ACC_g375h-label.nii.gz",
141
+ "image": "ACC_g375h/ACC_g375h-data.nii.gz"
142
+ },
143
+ {
144
+ "label": "ACC_g579h/ACC_g579h-label.nii.gz",
145
+ "image": "ACC_g579h/ACC_g579h-data.nii.gz"
146
+ },
147
+ {
148
+ "label": "acc_g054h/acc_g054h-label.nii.gz",
149
+ "image": "acc_g054h/acc_g054h-data.nii.gz"
150
+ },
151
+ {
152
+ "label": "ACC_g211h/ACC_g211h-label.nii.gz",
153
+ "image": "ACC_g211h/ACC_g211h-data.nii.gz"
154
+ },
155
+ {
156
+ "label": "ACC_g254h/ACC_g254h-label.nii.gz",
157
+ "image": "ACC_g254h/ACC_g254h-data.nii.gz"
158
+ },
159
+ {
160
+ "label": "ACC_g470h/ACC_g470h-label.nii.gz",
161
+ "image": "ACC_g470h/ACC_g470h-data.nii.gz"
162
+ },
163
+ {
164
+ "label": "acc_g072h/acc_g072h-label.nii.gz",
165
+ "image": "acc_g072h/acc_g072h-data.nii.gz"
166
+ },
167
+ {
168
+ "label": "ACC_g149h/ACC_g149h-label.nii.gz",
169
+ "image": "ACC_g149h/ACC_g149h-data.nii.gz"
170
+ },
171
+ {
172
+ "label": "ACC_g324h/ACC_g324h-label.nii.gz",
173
+ "image": "ACC_g324h/ACC_g324h-data.nii.gz"
174
+ },
175
+ {
176
+ "label": "ACC_g215h/ACC_g215h-label.nii.gz",
177
+ "image": "ACC_g215h/ACC_g215h-data.nii.gz"
178
+ },
179
+ {
180
+ "label": "ACC_g175h/ACC_g175h-label.nii.gz",
181
+ "image": "ACC_g175h/ACC_g175h-data.nii.gz"
182
+ },
183
+ {
184
+ "label": "ACC_g127h/ACC_g127h-label.nii.gz",
185
+ "image": "ACC_g127h/ACC_g127h-data.nii.gz"
186
+ },
187
+ {
188
+ "label": "ACC_g532h/ACC_g532h-label.nii.gz",
189
+ "image": "ACC_g532h/ACC_g532h-data.nii.gz"
190
+ },
191
+ {
192
+ "label": "acc_g005/acc_g005-label.nii.gz",
193
+ "image": "acc_g005/acc_g005-data.nii.gz"
194
+ },
195
+ {
196
+ "label": "ACC_g501h/ACC_g501h-label.nii.gz",
197
+ "image": "ACC_g501h/ACC_g501h-data.nii.gz"
198
+ },
199
+ {
200
+ "label": "ACC_g689h/ACC_g689h-label.nii.gz",
201
+ "image": "ACC_g689h/ACC_g689h-data.nii.gz"
202
+ },
203
+ {
204
+ "label": "ACC_g384h/ACC_g384h-label.nii.gz",
205
+ "image": "ACC_g384h/ACC_g384h-data.nii.gz"
206
+ },
207
+ {
208
+ "label": "ACC_g408h/ACC_g408h-label.nii.gz",
209
+ "image": "ACC_g408h/ACC_g408h-data.nii.gz"
210
+ },
211
+ {
212
+ "label": "ACC_g243h/ACC_g243h-label.nii.gz",
213
+ "image": "ACC_g243h/ACC_g243h-data.nii.gz"
214
+ },
215
+ {
216
+ "label": "acc_g063h/acc_g063h-label.nii.gz",
217
+ "image": "acc_g063h/acc_g063h-data.nii.gz"
218
+ },
219
+ {
220
+ "label": "acc_g034/acc_g034-label.nii.gz",
221
+ "image": "acc_g034/acc_g034-data.nii.gz"
222
+ },
223
+ {
224
+ "label": "acc_g088h/acc_g088h-label.nii.gz",
225
+ "image": "acc_g088h/acc_g088h-data.nii.gz"
226
+ },
227
+ {
228
+ "label": "acc_g014/acc_g014-label.nii.gz",
229
+ "image": "acc_g014/acc_g014-data.nii.gz"
230
+ },
231
+ {
232
+ "label": "ACC_g504h/ACC_g504h-label.nii.gz",
233
+ "image": "ACC_g504h/ACC_g504h-data.nii.gz"
234
+ },
235
+ {
236
+ "label": "acc_g067h/acc_g067h-label.nii.gz",
237
+ "image": "acc_g067h/acc_g067h-data.nii.gz"
238
+ },
239
+ {
240
+ "label": "acc_g028/acc_g028-label.nii.gz",
241
+ "image": "acc_g028/acc_g028-data.nii.gz"
242
+ },
243
+ {
244
+ "label": "ACC_g105h/ACC_g105h-label.nii.gz",
245
+ "image": "ACC_g105h/ACC_g105h-data.nii.gz"
246
+ },
247
+ {
248
+ "label": "ACC_g453h/ACC_g453h-label.nii.gz",
249
+ "image": "ACC_g453h/ACC_g453h-data.nii.gz"
250
+ },
251
+ {
252
+ "label": "acc_g052h/acc_g052h-label.nii.gz",
253
+ "image": "acc_g052h/acc_g052h-data.nii.gz"
254
+ },
255
+ {
256
+ "label": "ACC_g637h/ACC_g637h-label.nii.gz",
257
+ "image": "ACC_g637h/ACC_g637h-data.nii.gz"
258
+ },
259
+ {
260
+ "label": "ACC_g422h/ACC_g422h-label.nii.gz",
261
+ "image": "ACC_g422h/ACC_g422h-data.nii.gz"
262
+ },
263
+ {
264
+ "label": "ACC_g352h/ACC_g352h-label.nii.gz",
265
+ "image": "ACC_g352h/ACC_g352h-data.nii.gz"
266
+ },
267
+ {
268
+ "label": "acc_g071h/acc_g071h-label.nii.gz",
269
+ "image": "acc_g071h/acc_g071h-data.nii.gz"
270
+ },
271
+ {
272
+ "label": "ACC_g363h/ACC_g363h-label.nii.gz",
273
+ "image": "ACC_g363h/ACC_g363h-data.nii.gz"
274
+ },
275
+ {
276
+ "label": "ACC_g225h/ACC_g225h-label.nii.gz",
277
+ "image": "ACC_g225h/ACC_g225h-data.nii.gz"
278
+ },
279
+ {
280
+ "label": "ACC_g298h/ACC_g298h-label.nii.gz",
281
+ "image": "ACC_g298h/ACC_g298h-data.nii.gz"
282
+ },
283
+ {
284
+ "label": "ACC_g370h/ACC_g370h-label.nii.gz",
285
+ "image": "ACC_g370h/ACC_g370h-data.nii.gz"
286
+ },
287
+ {
288
+ "label": "ACC_g621h/ACC_g621h-label.nii.gz",
289
+ "image": "ACC_g621h/ACC_g621h-data.nii.gz"
290
+ },
291
+ {
292
+ "label": "ACC_g552h/ACC_g552h-label.nii.gz",
293
+ "image": "ACC_g552h/ACC_g552h-data.nii.gz"
294
+ },
295
+ {
296
+ "label": "ACC_g744h/ACC_g744h-label.nii.gz",
297
+ "image": "ACC_g744h/ACC_g744h-data.nii.gz"
298
+ },
299
+ {
300
+ "label": "ACC_g278h/ACC_g278h-label.nii.gz",
301
+ "image": "ACC_g278h/ACC_g278h-data.nii.gz"
302
+ },
303
+ {
304
+ "label": "ACC_g193h/ACC_g193h-label.nii.gz",
305
+ "image": "ACC_g193h/ACC_g193h-data.nii.gz"
306
+ },
307
+ {
308
+ "label": "ACC_g287h/ACC_g287h-label.nii.gz",
309
+ "image": "ACC_g287h/ACC_g287h-data.nii.gz"
310
+ },
311
+ {
312
+ "label": "ACC_g320h/ACC_g320h-label.nii.gz",
313
+ "image": "ACC_g320h/ACC_g320h-data.nii.gz"
314
+ },
315
+ {
316
+ "label": "ACC_g226h/ACC_g226h-label.nii.gz",
317
+ "image": "ACC_g226h/ACC_g226h-data.nii.gz"
318
+ },
319
+ {
320
+ "label": "acc_g085h/acc_g085h-label.nii.gz",
321
+ "image": "acc_g085h/acc_g085h-data.nii.gz"
322
+ },
323
+ {
324
+ "label": "ACC_g616h/ACC_g616h-label.nii.gz",
325
+ "image": "ACC_g616h/ACC_g616h-data.nii.gz"
326
+ },
327
+ {
328
+ "label": "ACC_g159h/ACC_g159h-label.nii.gz",
329
+ "image": "ACC_g159h/ACC_g159h-data.nii.gz"
330
+ },
331
+ {
332
+ "label": "ACC_g385h/ACC_g385h-label.nii.gz",
333
+ "image": "ACC_g385h/ACC_g385h-data.nii.gz"
334
+ },
335
+ {
336
+ "label": "ACC_g635h/ACC_g635h-label.nii.gz",
337
+ "image": "ACC_g635h/ACC_g635h-data.nii.gz"
338
+ },
339
+ {
340
+ "label": "ACC_g242h/ACC_g242h-label.nii.gz",
341
+ "image": "ACC_g242h/ACC_g242h-data.nii.gz"
342
+ },
343
+ {
344
+ "label": "ACC_g764h/ACC_g764h-label.nii.gz",
345
+ "image": "ACC_g764h/ACC_g764h-data.nii.gz"
346
+ },
347
+ {
348
+ "label": "acc_g059h/acc_g059h-label.nii.gz",
349
+ "image": "acc_g059h/acc_g059h-data.nii.gz"
350
+ },
351
+ {
352
+ "label": "acc_g035/acc_g035-label.nii.gz",
353
+ "image": "acc_g035/acc_g035-data.nii.gz"
354
+ },
355
+ {
356
+ "label": "acc_g041/acc_g041-label.nii.gz",
357
+ "image": "acc_g041/acc_g041-data.nii.gz"
358
+ },
359
+ {
360
+ "label": "ACC_g711h/ACC_g711h-label.nii.gz",
361
+ "image": "ACC_g711h/ACC_g711h-data.nii.gz"
362
+ },
363
+ {
364
+ "label": "ACC_g388h/ACC_g388h-label.nii.gz",
365
+ "image": "ACC_g388h/ACC_g388h-data.nii.gz"
366
+ },
367
+ {
368
+ "label": "ACC_g500h/ACC_g500h-label.nii.gz",
369
+ "image": "ACC_g500h/ACC_g500h-data.nii.gz"
370
+ },
371
+ {
372
+ "label": "ACC_g390h/ACC_g390h-label.nii.gz",
373
+ "image": "ACC_g390h/ACC_g390h-data.nii.gz"
374
+ },
375
+ {
376
+ "label": "ACC_g551h/ACC_g551h-label.nii.gz",
377
+ "image": "ACC_g551h/ACC_g551h-data.nii.gz"
378
+ },
379
+ {
380
+ "label": "acc_g053h/acc_g053h-label.nii.gz",
381
+ "image": "acc_g053h/acc_g053h-data.nii.gz"
382
+ },
383
+ {
384
+ "label": "ACC_g108h/ACC_g108h-label.nii.gz",
385
+ "image": "ACC_g108h/ACC_g108h-data.nii.gz"
386
+ },
387
+ {
388
+ "label": "acc_g057h/acc_g057h-label.nii.gz",
389
+ "image": "acc_g057h/acc_g057h-data.nii.gz"
390
+ },
391
+ {
392
+ "label": "ACC_g295h/ACC_g295h-label.nii.gz",
393
+ "image": "ACC_g295h/ACC_g295h-data.nii.gz"
394
+ },
395
+ {
396
+ "label": "ACC_g775h/ACC_g775h-label.nii.gz",
397
+ "image": "ACC_g775h/ACC_g775h-data.nii.gz"
398
+ },
399
+ {
400
+ "label": "ACC_g603h/ACC_g603h-label.nii.gz",
401
+ "image": "ACC_g603h/ACC_g603h-data.nii.gz"
402
+ },
403
+ {
404
+ "label": "acc_g045/acc_g045-label.nii.gz",
405
+ "image": "acc_g045/acc_g045-data.nii.gz"
406
+ },
407
+ {
408
+ "label": "ACC_g335h/ACC_g335h-label.nii.gz",
409
+ "image": "ACC_g335h/ACC_g335h-data.nii.gz"
410
+ },
411
+ {
412
+ "label": "ACC_g566h/ACC_g566h-label.nii.gz",
413
+ "image": "ACC_g566h/ACC_g566h-data.nii.gz"
414
+ },
415
+ {
416
+ "label": "ACC_g615h/ACC_g615h-label.nii.gz",
417
+ "image": "ACC_g615h/ACC_g615h-data.nii.gz"
418
+ },
419
+ {
420
+ "label": "ACC_g583h/ACC_g583h-label.nii.gz",
421
+ "image": "ACC_g583h/ACC_g583h-data.nii.gz"
422
+ },
423
+ {
424
+ "label": "acc_g038/acc_g038-label.nii.gz",
425
+ "image": "acc_g038/acc_g038-data.nii.gz"
426
+ },
427
+ {
428
+ "label": "ACC_g355h/ACC_g355h-label.nii.gz",
429
+ "image": "ACC_g355h/ACC_g355h-data.nii.gz"
430
+ },
431
+ {
432
+ "label": "ACC_g665h/ACC_g665h-label.nii.gz",
433
+ "image": "ACC_g665h/ACC_g665h-data.nii.gz"
434
+ },
435
+ {
436
+ "label": "ACC_g234h/ACC_g234h-label.nii.gz",
437
+ "image": "ACC_g234h/ACC_g234h-data.nii.gz"
438
+ },
439
+ {
440
+ "label": "ACC_g745h/ACC_g745h-label.nii.gz",
441
+ "image": "ACC_g745h/ACC_g745h-data.nii.gz"
442
+ },
443
+ {
444
+ "label": "ACC_g167h/ACC_g167h-label.nii.gz",
445
+ "image": "ACC_g167h/ACC_g167h-data.nii.gz"
446
+ },
447
+ {
448
+ "label": "ACC_g152h/ACC_g152h-label.nii.gz",
449
+ "image": "ACC_g152h/ACC_g152h-data.nii.gz"
450
+ },
451
+ {
452
+ "label": "ACC_g148h/ACC_g148h-label.nii.gz",
453
+ "image": "ACC_g148h/ACC_g148h-data.nii.gz"
454
+ },
455
+ {
456
+ "label": "acc_g011/acc_g011-label.nii.gz",
457
+ "image": "acc_g011/acc_g011-data.nii.gz"
458
+ },
459
+ {
460
+ "label": "ACC_g521h/ACC_g521h-label.nii.gz",
461
+ "image": "ACC_g521h/ACC_g521h-data.nii.gz"
462
+ },
463
+ {
464
+ "label": "ACC_g130h/ACC_g130h-label.nii.gz",
465
+ "image": "ACC_g130h/ACC_g130h-data.nii.gz"
466
+ },
467
+ {
468
+ "label": "acc_g096h/acc_g096h-label.nii.gz",
469
+ "image": "acc_g096h/acc_g096h-data.nii.gz"
470
+ },
471
+ {
472
+ "label": "acc_g050h/acc_g050h-label.nii.gz",
473
+ "image": "acc_g050h/acc_g050h-data.nii.gz"
474
+ },
475
+ {
476
+ "label": "ACC_g643h/ACC_g643h-label.nii.gz",
477
+ "image": "ACC_g643h/ACC_g643h-data.nii.gz"
478
+ },
479
+ {
480
+ "label": "ACC_g534h/ACC_g534h-label.nii.gz",
481
+ "image": "ACC_g534h/ACC_g534h-data.nii.gz"
482
+ },
483
+ {
484
+ "label": "ACC_g111h/ACC_g111h-label.nii.gz",
485
+ "image": "ACC_g111h/ACC_g111h-data.nii.gz"
486
+ },
487
+ {
488
+ "label": "ACC_g202h/ACC_g202h-label.nii.gz",
489
+ "image": "ACC_g202h/ACC_g202h-data.nii.gz"
490
+ },
491
+ {
492
+ "label": "ACC_g139h/ACC_g139h-label.nii.gz",
493
+ "image": "ACC_g139h/ACC_g139h-data.nii.gz"
494
+ },
495
+ {
496
+ "label": "ACC_g494h/ACC_g494h-label.nii.gz",
497
+ "image": "ACC_g494h/ACC_g494h-data.nii.gz"
498
+ },
499
+ {
500
+ "label": "ACC_g693h/ACC_g693h-label.nii.gz",
501
+ "image": "ACC_g693h/ACC_g693h-data.nii.gz"
502
+ },
503
+ {
504
+ "label": "ACC_g315h/ACC_g315h-label.nii.gz",
505
+ "image": "ACC_g315h/ACC_g315h-data.nii.gz"
506
+ },
507
+ {
508
+ "label": "ACC_g690h/ACC_g690h-label.nii.gz",
509
+ "image": "ACC_g690h/ACC_g690h-data.nii.gz"
510
+ },
511
+ {
512
+ "label": "ACC_g380h/ACC_g380h-label.nii.gz",
513
+ "image": "ACC_g380h/ACC_g380h-data.nii.gz"
514
+ },
515
+ {
516
+ "label": "ACC_g535h/ACC_g535h-label.nii.gz",
517
+ "image": "ACC_g535h/ACC_g535h-data.nii.gz"
518
+ },
519
+ {
520
+ "label": "ACC_g460h/ACC_g460h-label.nii.gz",
521
+ "image": "ACC_g460h/ACC_g460h-data.nii.gz"
522
+ },
523
+ {
524
+ "label": "ACC_g340h/ACC_g340h-label.nii.gz",
525
+ "image": "ACC_g340h/ACC_g340h-data.nii.gz"
526
+ },
527
+ {
528
+ "label": "ACC_g691h/ACC_g691h-label.nii.gz",
529
+ "image": "ACC_g691h/ACC_g691h-data.nii.gz"
530
+ },
531
+ {
532
+ "label": "acc_g064h/acc_g064h-label.nii.gz",
533
+ "image": "acc_g064h/acc_g064h-data.nii.gz"
534
+ },
535
+ {
536
+ "label": "ACC_g291h/ACC_g291h-label.nii.gz",
537
+ "image": "ACC_g291h/ACC_g291h-data.nii.gz"
538
+ },
539
+ {
540
+ "label": "ACC_g206h/ACC_g206h-label.nii.gz",
541
+ "image": "ACC_g206h/ACC_g206h-data.nii.gz"
542
+ },
543
+ {
544
+ "label": "ACC_g331h/ACC_g331h-label.nii.gz",
545
+ "image": "ACC_g331h/ACC_g331h-data.nii.gz"
546
+ },
547
+ {
548
+ "label": "ACC_g274h/ACC_g274h-label.nii.gz",
549
+ "image": "ACC_g274h/ACC_g274h-data.nii.gz"
550
+ },
551
+ {
552
+ "label": "ACC_g649h/ACC_g649h-label.nii.gz",
553
+ "image": "ACC_g649h/ACC_g649h-data.nii.gz"
554
+ },
555
+ {
556
+ "label": "ACC_g736h/ACC_g736h-label.nii.gz",
557
+ "image": "ACC_g736h/ACC_g736h-data.nii.gz"
558
+ },
559
+ {
560
+ "label": "ACC_g761h/ACC_g761h-label.nii.gz",
561
+ "image": "ACC_g761h/ACC_g761h-data.nii.gz"
562
+ },
563
+ {
564
+ "label": "ACC_g289h/ACC_g289h-label.nii.gz",
565
+ "image": "ACC_g289h/ACC_g289h-data.nii.gz"
566
+ },
567
+ {
568
+ "label": "ACC_g448h/ACC_g448h-label.nii.gz",
569
+ "image": "ACC_g448h/ACC_g448h-data.nii.gz"
570
+ },
571
+ {
572
+ "label": "ACC_g310h/ACC_g310h-label.nii.gz",
573
+ "image": "ACC_g310h/ACC_g310h-data.nii.gz"
574
+ },
575
+ {
576
+ "label": "acc_g009/acc_g009-label.nii.gz",
577
+ "image": "acc_g009/acc_g009-data.nii.gz"
578
+ },
579
+ {
580
+ "label": "ACC_g247h/ACC_g247h-label.nii.gz",
581
+ "image": "ACC_g247h/ACC_g247h-data.nii.gz"
582
+ },
583
+ {
584
+ "label": "ACC_g773h/ACC_g773h-label.nii.gz",
585
+ "image": "ACC_g773h/ACC_g773h-data.nii.gz"
586
+ },
587
+ {
588
+ "label": "ACC_g465h/ACC_g465h-label.nii.gz",
589
+ "image": "ACC_g465h/ACC_g465h-data.nii.gz"
590
+ },
591
+ {
592
+ "label": "ACC_g196h/ACC_g196h-label.nii.gz",
593
+ "image": "ACC_g196h/ACC_g196h-data.nii.gz"
594
+ },
595
+ {
596
+ "label": "acc_g068h/acc_g068h-label.nii.gz",
597
+ "image": "acc_g068h/acc_g068h-data.nii.gz"
598
+ },
599
+ {
600
+ "label": "ACC_g409h/ACC_g409h-label.nii.gz",
601
+ "image": "ACC_g409h/ACC_g409h-data.nii.gz"
602
+ },
603
+ {
604
+ "label": "acc_g006/acc_g006-label.nii.gz",
605
+ "image": "acc_g006/acc_g006-data.nii.gz"
606
+ },
607
+ {
608
+ "label": "ACC_g160h/ACC_g160h-label.nii.gz",
609
+ "image": "ACC_g160h/ACC_g160h-data.nii.gz"
610
+ },
611
+ {
612
+ "label": "acc_g027/acc_g027-label.nii.gz",
613
+ "image": "acc_g027/acc_g027-data.nii.gz"
614
+ },
615
+ {
616
+ "label": "ACC_g326h/ACC_g326h-label.nii.gz",
617
+ "image": "ACC_g326h/ACC_g326h-data.nii.gz"
618
+ },
619
+ {
620
+ "label": "ACC_g170h/ACC_g170h-label.nii.gz",
621
+ "image": "ACC_g170h/ACC_g170h-data.nii.gz"
622
+ },
623
+ {
624
+ "label": "ACC_g269h/ACC_g269h-label.nii.gz",
625
+ "image": "ACC_g269h/ACC_g269h-data.nii.gz"
626
+ },
627
+ {
628
+ "label": "ACC_g468h/ACC_g468h-label.nii.gz",
629
+ "image": "ACC_g468h/ACC_g468h-data.nii.gz"
630
+ },
631
+ {
632
+ "label": "ACC_g512h/ACC_g512h-label.nii.gz",
633
+ "image": "ACC_g512h/ACC_g512h-data.nii.gz"
634
+ },
635
+ {
636
+ "label": "ACC_g563h/ACC_g563h-label.nii.gz",
637
+ "image": "ACC_g563h/ACC_g563h-data.nii.gz"
638
+ },
639
+ {
640
+ "label": "acc_g055h/acc_g055h-label.nii.gz",
641
+ "image": "acc_g055h/acc_g055h-data.nii.gz"
642
+ },
643
+ {
644
+ "label": "acc_g086h/acc_g086h-label.nii.gz",
645
+ "image": "acc_g086h/acc_g086h-data.nii.gz"
646
+ },
647
+ {
648
+ "label": "ACC_g762h/ACC_g762h-label.nii.gz",
649
+ "image": "ACC_g762h/ACC_g762h-data.nii.gz"
650
+ },
651
+ {
652
+ "label": "ACC_g197h/ACC_g197h-label.nii.gz",
653
+ "image": "ACC_g197h/ACC_g197h-data.nii.gz"
654
+ },
655
+ {
656
+ "label": "ACC_g674h/ACC_g674h-label.nii.gz",
657
+ "image": "ACC_g674h/ACC_g674h-data.nii.gz"
658
+ },
659
+ {
660
+ "label": "ACC_g393h/ACC_g393h-label.nii.gz",
661
+ "image": "ACC_g393h/ACC_g393h-data.nii.gz"
662
+ },
663
+ {
664
+ "label": "ACC_g578h/ACC_g578h-label.nii.gz",
665
+ "image": "ACC_g578h/ACC_g578h-data.nii.gz"
666
+ },
667
+ {
668
+ "label": "ACC_g601h/ACC_g601h-label.nii.gz",
669
+ "image": "ACC_g601h/ACC_g601h-data.nii.gz"
670
+ },
671
+ {
672
+ "label": "ACC_g586h/ACC_g586h-label.nii.gz",
673
+ "image": "ACC_g586h/ACC_g586h-data.nii.gz"
674
+ },
675
+ {
676
+ "label": "ACC_g626h/ACC_g626h-label.nii.gz",
677
+ "image": "ACC_g626h/ACC_g626h-data.nii.gz"
678
+ },
679
+ {
680
+ "label": "acc_g008/acc_g008-label.nii.gz",
681
+ "image": "acc_g008/acc_g008-data.nii.gz"
682
+ },
683
+ {
684
+ "label": "acc_g077h/acc_g077h-label.nii.gz",
685
+ "image": "acc_g077h/acc_g077h-data.nii.gz"
686
+ },
687
+ {
688
+ "label": "ACC_g365h/ACC_g365h-label.nii.gz",
689
+ "image": "ACC_g365h/ACC_g365h-data.nii.gz"
690
+ },
691
+ {
692
+ "label": "acc_g036/acc_g036-label.nii.gz",
693
+ "image": "acc_g036/acc_g036-data.nii.gz"
694
+ },
695
+ {
696
+ "label": "ACC_g155h/ACC_g155h-label.nii.gz",
697
+ "image": "ACC_g155h/ACC_g155h-data.nii.gz"
698
+ },
699
+ {
700
+ "label": "acc_g033/acc_g033-label.nii.gz",
701
+ "image": "acc_g033/acc_g033-data.nii.gz"
702
+ },
703
+ {
704
+ "label": "ACC_g751h/ACC_g751h-label.nii.gz",
705
+ "image": "ACC_g751h/ACC_g751h-data.nii.gz"
706
+ },
707
+ {
708
+ "label": "ACC_g201h/ACC_g201h-label.nii.gz",
709
+ "image": "ACC_g201h/ACC_g201h-data.nii.gz"
710
+ },
711
+ {
712
+ "label": "ACC_g491h/ACC_g491h-label.nii.gz",
713
+ "image": "ACC_g491h/ACC_g491h-data.nii.gz"
714
+ },
715
+ {
716
+ "label": "ACC_g109h/ACC_g109h-label.nii.gz",
717
+ "image": "ACC_g109h/ACC_g109h-data.nii.gz"
718
+ },
719
+ {
720
+ "label": "ACC_g219h/ACC_g219h-label.nii.gz",
721
+ "image": "ACC_g219h/ACC_g219h-data.nii.gz"
722
+ },
723
+ {
724
+ "label": "ACC_g189h/ACC_g189h-label.nii.gz",
725
+ "image": "ACC_g189h/ACC_g189h-data.nii.gz"
726
+ },
727
+ {
728
+ "label": "ACC_g585h/ACC_g585h-label.nii.gz",
729
+ "image": "ACC_g585h/ACC_g585h-data.nii.gz"
730
+ },
731
+ {
732
+ "label": "ACC_g301h/ACC_g301h-label.nii.gz",
733
+ "image": "ACC_g301h/ACC_g301h-data.nii.gz"
734
+ },
735
+ {
736
+ "label": "ACC_g346h/ACC_g346h-label.nii.gz",
737
+ "image": "ACC_g346h/ACC_g346h-data.nii.gz"
738
+ },
739
+ {
740
+ "label": "ACC_g241h/ACC_g241h-label.nii.gz",
741
+ "image": "ACC_g241h/ACC_g241h-data.nii.gz"
742
+ },
743
+ {
744
+ "label": "acc_g049h/acc_g049h-label.nii.gz",
745
+ "image": "acc_g049h/acc_g049h-data.nii.gz"
746
+ },
747
+ {
748
+ "label": "ACC_g680h/ACC_g680h-label.nii.gz",
749
+ "image": "ACC_g680h/ACC_g680h-data.nii.gz"
750
+ },
751
+ {
752
+ "label": "ACC_g592h/ACC_g592h-label.nii.gz",
753
+ "image": "ACC_g592h/ACC_g592h-data.nii.gz"
754
+ },
755
+ {
756
+ "label": "ACC_g192h/ACC_g192h-label.nii.gz",
757
+ "image": "ACC_g192h/ACC_g192h-data.nii.gz"
758
+ },
759
+ {
760
+ "label": "ACC_g597h/ACC_g597h-label.nii.gz",
761
+ "image": "ACC_g597h/ACC_g597h-data.nii.gz"
762
+ },
763
+ {
764
+ "label": "ACC_g681h/ACC_g681h-label.nii.gz",
765
+ "image": "ACC_g681h/ACC_g681h-data.nii.gz"
766
+ },
767
+ {
768
+ "label": "ACC_g673h/ACC_g673h-label.nii.gz",
769
+ "image": "ACC_g673h/ACC_g673h-data.nii.gz"
770
+ },
771
+ {
772
+ "label": "acc_g026/acc_g026-label.nii.gz",
773
+ "image": "acc_g026/acc_g026-data.nii.gz"
774
+ },
775
+ {
776
+ "label": "ACC_g161h/ACC_g161h-label.nii.gz",
777
+ "image": "ACC_g161h/ACC_g161h-data.nii.gz"
778
+ },
779
+ {
780
+ "label": "ACC_g235h/ACC_g235h-label.nii.gz",
781
+ "image": "ACC_g235h/ACC_g235h-data.nii.gz"
782
+ },
783
+ {
784
+ "label": "ACC_g288h/ACC_g288h-label.nii.gz",
785
+ "image": "ACC_g288h/ACC_g288h-data.nii.gz"
786
+ },
787
+ {
788
+ "label": "ACC_g332h/ACC_g332h-label.nii.gz",
789
+ "image": "ACC_g332h/ACC_g332h-data.nii.gz"
790
+ },
791
+ {
792
+ "label": "ACC_g118h/ACC_g118h-label.nii.gz",
793
+ "image": "ACC_g118h/ACC_g118h-data.nii.gz"
794
+ },
795
+ {
796
+ "label": "ACC_g684h/ACC_g684h-label.nii.gz",
797
+ "image": "ACC_g684h/ACC_g684h-data.nii.gz"
798
+ },
799
+ {
800
+ "label": "ACC_g459h/ACC_g459h-label.nii.gz",
801
+ "image": "ACC_g459h/ACC_g459h-data.nii.gz"
802
+ },
803
+ {
804
+ "label": "acc_g094h/acc_g094h-label.nii.gz",
805
+ "image": "acc_g094h/acc_g094h-data.nii.gz"
806
+ },
807
+ {
808
+ "label": "ACC_g297h/ACC_g297h-label.nii.gz",
809
+ "image": "ACC_g297h/ACC_g297h-data.nii.gz"
810
+ },
811
+ {
812
+ "label": "ACC_g671h/ACC_g671h-label.nii.gz",
813
+ "image": "ACC_g671h/ACC_g671h-data.nii.gz"
814
+ },
815
+ {
816
+ "label": "ACC_g137h/ACC_g137h-label.nii.gz",
817
+ "image": "ACC_g137h/ACC_g137h-data.nii.gz"
818
+ },
819
+ {
820
+ "label": "ACC_g343h/ACC_g343h-label.nii.gz",
821
+ "image": "ACC_g343h/ACC_g343h-data.nii.gz"
822
+ },
823
+ {
824
+ "label": "ACC_g469h/ACC_g469h-label.nii.gz",
825
+ "image": "ACC_g469h/ACC_g469h-data.nii.gz"
826
+ },
827
+ {
828
+ "label": "ACC_g116h/ACC_g116h-label.nii.gz",
829
+ "image": "ACC_g116h/ACC_g116h-data.nii.gz"
830
+ },
831
+ {
832
+ "label": "ACC_g239h/ACC_g239h-label.nii.gz",
833
+ "image": "ACC_g239h/ACC_g239h-data.nii.gz"
834
+ },
835
+ {
836
+ "label": "ACC_g617h/ACC_g617h-label.nii.gz",
837
+ "image": "ACC_g617h/ACC_g617h-data.nii.gz"
838
+ },
839
+ {
840
+ "label": "ACC_g611h/ACC_g611h-label.nii.gz",
841
+ "image": "ACC_g611h/ACC_g611h-data.nii.gz"
842
+ },
843
+ {
844
+ "label": "ACC_g527h/ACC_g527h-label.nii.gz",
845
+ "image": "ACC_g527h/ACC_g527h-data.nii.gz"
846
+ },
847
+ {
848
+ "label": "ACC_g570h/ACC_g570h-label.nii.gz",
849
+ "image": "ACC_g570h/ACC_g570h-data.nii.gz"
850
+ },
851
+ {
852
+ "label": "ACC_g606h/ACC_g606h-label.nii.gz",
853
+ "image": "ACC_g606h/ACC_g606h-data.nii.gz"
854
+ },
855
+ {
856
+ "label": "ACC_g776h/ACC_g776h-label.nii.gz",
857
+ "image": "ACC_g776h/ACC_g776h-data.nii.gz"
858
+ },
859
+ {
860
+ "label": "ACC_g277h/ACC_g277h-label.nii.gz",
861
+ "image": "ACC_g277h/ACC_g277h-data.nii.gz"
862
+ },
863
+ {
864
+ "label": "ACC_g360h/ACC_g360h-label.nii.gz",
865
+ "image": "ACC_g360h/ACC_g360h-data.nii.gz"
866
+ },
867
+ {
868
+ "label": "ACC_g342h/ACC_g342h-label.nii.gz",
869
+ "image": "ACC_g342h/ACC_g342h-data.nii.gz"
870
+ },
871
+ {
872
+ "label": "ACC_g203h/ACC_g203h-label.nii.gz",
873
+ "image": "ACC_g203h/ACC_g203h-data.nii.gz"
874
+ },
875
+ {
876
+ "label": "ACC_g101h/ACC_g101h-label.nii.gz",
877
+ "image": "ACC_g101h/ACC_g101h-data.nii.gz"
878
+ },
879
+ {
880
+ "label": "ACC_g178h/ACC_g178h-label.nii.gz",
881
+ "image": "ACC_g178h/ACC_g178h-data.nii.gz"
882
+ },
883
+ {
884
+ "label": "ACC_g687h/ACC_g687h-label.nii.gz",
885
+ "image": "ACC_g687h/ACC_g687h-data.nii.gz"
886
+ },
887
+ {
888
+ "label": "ACC_g542h/ACC_g542h-label.nii.gz",
889
+ "image": "ACC_g542h/ACC_g542h-data.nii.gz"
890
+ },
891
+ {
892
+ "label": "ACC_g171h/ACC_g171h-label.nii.gz",
893
+ "image": "ACC_g171h/ACC_g171h-data.nii.gz"
894
+ },
895
+ {
896
+ "label": "ACC_g098h/ACC_g098h-label.nii.gz",
897
+ "image": "ACC_g098h/ACC_g098h-data.nii.gz"
898
+ },
899
+ {
900
+ "label": "ACC_g120h/ACC_g120h-label.nii.gz",
901
+ "image": "ACC_g120h/ACC_g120h-data.nii.gz"
902
+ },
903
+ {
904
+ "label": "ACC_g622h/ACC_g622h-label.nii.gz",
905
+ "image": "ACC_g622h/ACC_g622h-data.nii.gz"
906
+ },
907
+ {
908
+ "label": "acc_g058h/acc_g058h-label.nii.gz",
909
+ "image": "acc_g058h/acc_g058h-data.nii.gz"
910
+ },
911
+ {
912
+ "label": "ACC_g704h/ACC_g704h-label.nii.gz",
913
+ "image": "ACC_g704h/ACC_g704h-data.nii.gz"
914
+ },
915
+ {
916
+ "label": "ACC_g132h/ACC_g132h-label.nii.gz",
917
+ "image": "ACC_g132h/ACC_g132h-data.nii.gz"
918
+ },
919
+ {
920
+ "label": "ACC_g441h/ACC_g441h-label.nii.gz",
921
+ "image": "ACC_g441h/ACC_g441h-data.nii.gz"
922
+ },
923
+ {
924
+ "label": "ACC_g447h/ACC_g447h-label.nii.gz",
925
+ "image": "ACC_g447h/ACC_g447h-data.nii.gz"
926
+ },
927
+ {
928
+ "label": "ACC_g158h/ACC_g158h-label.nii.gz",
929
+ "image": "ACC_g158h/ACC_g158h-data.nii.gz"
930
+ },
931
+ {
932
+ "label": "ACC_g325h/ACC_g325h-label.nii.gz",
933
+ "image": "ACC_g325h/ACC_g325h-data.nii.gz"
934
+ },
935
+ {
936
+ "label": "ACC_g572h/ACC_g572h-label.nii.gz",
937
+ "image": "ACC_g572h/ACC_g572h-data.nii.gz"
938
+ },
939
+ {
940
+ "label": "ACC_g596h/ACC_g596h-label.nii.gz",
941
+ "image": "ACC_g596h/ACC_g596h-data.nii.gz"
942
+ },
943
+ {
944
+ "label": "ACC_g628h/ACC_g628h-label.nii.gz",
945
+ "image": "ACC_g628h/ACC_g628h-data.nii.gz"
946
+ },
947
+ {
948
+ "label": "acc_g060h/acc_g060h-label.nii.gz",
949
+ "image": "acc_g060h/acc_g060h-data.nii.gz"
950
+ },
951
+ {
952
+ "label": "ACC_g367h/ACC_g367h-label.nii.gz",
953
+ "image": "ACC_g367h/ACC_g367h-data.nii.gz"
954
+ },
955
+ {
956
+ "label": "ACC_g207h/ACC_g207h-label.nii.gz",
957
+ "image": "ACC_g207h/ACC_g207h-data.nii.gz"
958
+ },
959
+ {
960
+ "label": "acc_g043/acc_g043-label.nii.gz",
961
+ "image": "acc_g043/acc_g043-data.nii.gz"
962
+ },
963
+ {
964
+ "label": "ACC_g373h/ACC_g373h-label.nii.gz",
965
+ "image": "ACC_g373h/ACC_g373h-data.nii.gz"
966
+ },
967
+ {
968
+ "label": "acc_g062h/acc_g062h-label.nii.gz",
969
+ "image": "acc_g062h/acc_g062h-data.nii.gz"
970
+ },
971
+ {
972
+ "label": "ACC_g629h/ACC_g629h-label.nii.gz",
973
+ "image": "ACC_g629h/ACC_g629h-data.nii.gz"
974
+ },
975
+ {
976
+ "label": "ACC_g357h/ACC_g357h-label.nii.gz",
977
+ "image": "ACC_g357h/ACC_g357h-data.nii.gz"
978
+ },
979
+ {
980
+ "label": "ACC_g488h/ACC_g488h-label.nii.gz",
981
+ "image": "ACC_g488h/ACC_g488h-data.nii.gz"
982
+ },
983
+ {
984
+ "label": "ACC_g640h/ACC_g640h-label.nii.gz",
985
+ "image": "ACC_g640h/ACC_g640h-data.nii.gz"
986
+ },
987
+ {
988
+ "label": "ACC_g708h/ACC_g708h-label.nii.gz",
989
+ "image": "ACC_g708h/ACC_g708h-data.nii.gz"
990
+ },
991
+ {
992
+ "label": "ACC_g753h/ACC_g753h-label.nii.gz",
993
+ "image": "ACC_g753h/ACC_g753h-data.nii.gz"
994
+ },
995
+ {
996
+ "label": "ACC_g200h/ACC_g200h-label.nii.gz",
997
+ "image": "ACC_g200h/ACC_g200h-data.nii.gz"
998
+ },
999
+ {
1000
+ "label": "ACC_g743h/ACC_g743h-label.nii.gz",
1001
+ "image": "ACC_g743h/ACC_g743h-data.nii.gz"
1002
+ },
1003
+ {
1004
+ "label": "acc_g082h/acc_g082h-label.nii.gz",
1005
+ "image": "acc_g082h/acc_g082h-data.nii.gz"
1006
+ },
1007
+ {
1008
+ "label": "ACC_g391h/ACC_g391h-label.nii.gz",
1009
+ "image": "ACC_g391h/ACC_g391h-data.nii.gz"
1010
+ },
1011
+ {
1012
+ "label": "ACC_g218h/ACC_g218h-label.nii.gz",
1013
+ "image": "ACC_g218h/ACC_g218h-data.nii.gz"
1014
+ },
1015
+ {
1016
+ "label": "ACC_g664h/ACC_g664h-label.nii.gz",
1017
+ "image": "ACC_g664h/ACC_g664h-data.nii.gz"
1018
+ },
1019
+ {
1020
+ "label": "acc_g012/acc_g012-label.nii.gz",
1021
+ "image": "acc_g012/acc_g012-data.nii.gz"
1022
+ },
1023
+ {
1024
+ "label": "ACC_g300h/ACC_g300h-label.nii.gz",
1025
+ "image": "ACC_g300h/ACC_g300h-data.nii.gz"
1026
+ },
1027
+ {
1028
+ "label": "ACC_g505h/ACC_g505h-label.nii.gz",
1029
+ "image": "ACC_g505h/ACC_g505h-data.nii.gz"
1030
+ },
1031
+ {
1032
+ "label": "acc_g015/acc_g015-label.nii.gz",
1033
+ "image": "acc_g015/acc_g015-data.nii.gz"
1034
+ },
1035
+ {
1036
+ "label": "acc_g020/acc_g020-label.nii.gz",
1037
+ "image": "acc_g020/acc_g020-data.nii.gz"
1038
+ },
1039
+ {
1040
+ "label": "ACC_g431h/ACC_g431h-label.nii.gz",
1041
+ "image": "ACC_g431h/ACC_g431h-data.nii.gz"
1042
+ },
1043
+ {
1044
+ "label": "acc_g010/acc_g010-label.nii.gz",
1045
+ "image": "acc_g010/acc_g010-data.nii.gz"
1046
+ },
1047
+ {
1048
+ "label": "acc_g044/acc_g044-label.nii.gz",
1049
+ "image": "acc_g044/acc_g044-data.nii.gz"
1050
+ },
1051
+ {
1052
+ "label": "ACC_g273h/ACC_g273h-label.nii.gz",
1053
+ "image": "ACC_g273h/ACC_g273h-data.nii.gz"
1054
+ },
1055
+ {
1056
+ "label": "acc_g025/acc_g025-label.nii.gz",
1057
+ "image": "acc_g025/acc_g025-data.nii.gz"
1058
+ },
1059
+ {
1060
+ "label": "ACC_g334h/ACC_g334h-label.nii.gz",
1061
+ "image": "ACC_g334h/ACC_g334h-data.nii.gz"
1062
+ },
1063
+ {
1064
+ "label": "ACC_g421h/ACC_g421h-label.nii.gz",
1065
+ "image": "ACC_g421h/ACC_g421h-data.nii.gz"
1066
+ },
1067
+ {
1068
+ "label": "acc_g069h/acc_g069h-label.nii.gz",
1069
+ "image": "acc_g069h/acc_g069h-data.nii.gz"
1070
+ },
1071
+ {
1072
+ "label": "ACC_g233h/ACC_g233h-label.nii.gz",
1073
+ "image": "ACC_g233h/ACC_g233h-data.nii.gz"
1074
+ },
1075
+ {
1076
+ "label": "acc_g090h/acc_g090h-label.nii.gz",
1077
+ "image": "acc_g090h/acc_g090h-data.nii.gz"
1078
+ },
1079
+ {
1080
+ "label": "ACC_g154h/ACC_g154h-label.nii.gz",
1081
+ "image": "ACC_g154h/ACC_g154h-data.nii.gz"
1082
+ },
1083
+ {
1084
+ "label": "acc_g002/acc_g002-label.nii.gz",
1085
+ "image": "acc_g002/acc_g002-data.nii.gz"
1086
+ },
1087
+ {
1088
+ "label": "ACC_g100h/ACC_g100h-label.nii.gz",
1089
+ "image": "ACC_g100h/ACC_g100h-data.nii.gz"
1090
+ },
1091
+ {
1092
+ "label": "ACC_g699h/ACC_g699h-label.nii.gz",
1093
+ "image": "ACC_g699h/ACC_g699h-data.nii.gz"
1094
+ },
1095
+ {
1096
+ "label": "ACC_g688h/ACC_g688h-label.nii.gz",
1097
+ "image": "ACC_g688h/ACC_g688h-data.nii.gz"
1098
+ },
1099
+ {
1100
+ "label": "ACC_g204h/ACC_g204h-label.nii.gz",
1101
+ "image": "ACC_g204h/ACC_g204h-data.nii.gz"
1102
+ },
1103
+ {
1104
+ "label": "ACC_g495h/ACC_g495h-label.nii.gz",
1105
+ "image": "ACC_g495h/ACC_g495h-data.nii.gz"
1106
+ },
1107
+ {
1108
+ "label": "acc_g073h/acc_g073h-label.nii.gz",
1109
+ "image": "acc_g073h/acc_g073h-data.nii.gz"
1110
+ },
1111
+ {
1112
+ "label": "ACC_g589h/ACC_g589h-label.nii.gz",
1113
+ "image": "ACC_g589h/ACC_g589h-data.nii.gz"
1114
+ },
1115
+ {
1116
+ "label": "ACC_g763h/ACC_g763h-label.nii.gz",
1117
+ "image": "ACC_g763h/ACC_g763h-data.nii.gz"
1118
+ },
1119
+ {
1120
+ "label": "acc_g001/acc_g001-label.nii.gz",
1121
+ "image": "acc_g001/acc_g001-data.nii.gz"
1122
+ },
1123
+ {
1124
+ "label": "ACC_g290h/ACC_g290h-label.nii.gz",
1125
+ "image": "ACC_g290h/ACC_g290h-data.nii.gz"
1126
+ },
1127
+ {
1128
+ "label": "ACC_g530h/ACC_g530h-label.nii.gz",
1129
+ "image": "ACC_g530h/ACC_g530h-data.nii.gz"
1130
+ },
1131
+ {
1132
+ "label": "ACC_g144h/ACC_g144h-label.nii.gz",
1133
+ "image": "ACC_g144h/ACC_g144h-data.nii.gz"
1134
+ },
1135
+ {
1136
+ "label": "ACC_g213h/ACC_g213h-label.nii.gz",
1137
+ "image": "ACC_g213h/ACC_g213h-data.nii.gz"
1138
+ },
1139
+ {
1140
+ "label": "ACC_g164h/ACC_g164h-label.nii.gz",
1141
+ "image": "ACC_g164h/ACC_g164h-data.nii.gz"
1142
+ },
1143
+ {
1144
+ "label": "ACC_g317h/ACC_g317h-label.nii.gz",
1145
+ "image": "ACC_g317h/ACC_g317h-data.nii.gz"
1146
+ },
1147
+ {
1148
+ "label": "ACC_g293h/ACC_g293h-label.nii.gz",
1149
+ "image": "ACC_g293h/ACC_g293h-data.nii.gz"
1150
+ },
1151
+ {
1152
+ "label": "ACC_g268h/ACC_g268h-label.nii.gz",
1153
+ "image": "ACC_g268h/ACC_g268h-data.nii.gz"
1154
+ },
1155
+ {
1156
+ "label": "ACC_g531h/ACC_g531h-label.nii.gz",
1157
+ "image": "ACC_g531h/ACC_g531h-data.nii.gz"
1158
+ },
1159
+ {
1160
+ "label": "ACC_g508h/ACC_g508h-label.nii.gz",
1161
+ "image": "ACC_g508h/ACC_g508h-data.nii.gz"
1162
+ },
1163
+ {
1164
+ "label": "ACC_g180h/ACC_g180h-label.nii.gz",
1165
+ "image": "ACC_g180h/ACC_g180h-data.nii.gz"
1166
+ },
1167
+ {
1168
+ "label": "ACC_g311h/ACC_g311h-label.nii.gz",
1169
+ "image": "ACC_g311h/ACC_g311h-data.nii.gz"
1170
+ },
1171
+ {
1172
+ "label": "ACC_g282h/ACC_g282h-label.nii.gz",
1173
+ "image": "ACC_g282h/ACC_g282h-data.nii.gz"
1174
+ },
1175
+ {
1176
+ "label": "ACC_g232h/ACC_g232h-label.nii.gz",
1177
+ "image": "ACC_g232h/ACC_g232h-data.nii.gz"
1178
+ },
1179
+ {
1180
+ "label": "ACC_g119h/ACC_g119h-label.nii.gz",
1181
+ "image": "ACC_g119h/ACC_g119h-data.nii.gz"
1182
+ },
1183
+ {
1184
+ "label": "ACC_g536h/ACC_g536h-label.nii.gz",
1185
+ "image": "ACC_g536h/ACC_g536h-data.nii.gz"
1186
+ },
1187
+ {
1188
+ "label": "ACC_g270h/ACC_g270h-label.nii.gz",
1189
+ "image": "ACC_g270h/ACC_g270h-data.nii.gz"
1190
+ },
1191
+ {
1192
+ "label": "ACC_g739h/ACC_g739h-label.nii.gz",
1193
+ "image": "ACC_g739h/ACC_g739h-data.nii.gz"
1194
+ },
1195
+ {
1196
+ "label": "acc_g047/acc_g047-label.nii.gz",
1197
+ "image": "acc_g047/acc_g047-data.nii.gz"
1198
+ },
1199
+ {
1200
+ "label": "acc_g029/acc_g029-label.nii.gz",
1201
+ "image": "acc_g029/acc_g029-data.nii.gz"
1202
+ }
1203
+ ],
1204
+ "validation": [
1205
+ {
1206
+ "label": "ACC_g115h/ACC_g115h-label.nii.gz",
1207
+ "image": "ACC_g115h/ACC_g115h-data.nii.gz"
1208
+ },
1209
+ {
1210
+ "label": "ACC_g568h/ACC_g568h-label.nii.gz",
1211
+ "image": "ACC_g568h/ACC_g568h-data.nii.gz"
1212
+ },
1213
+ {
1214
+ "label": "acc_g042/acc_g042-label.nii.gz",
1215
+ "image": "acc_g042/acc_g042-data.nii.gz"
1216
+ },
1217
+ {
1218
+ "label": "ACC_g541h/ACC_g541h-label.nii.gz",
1219
+ "image": "ACC_g541h/ACC_g541h-data.nii.gz"
1220
+ },
1221
+ {
1222
+ "label": "ACC_g245h/ACC_g245h-label.nii.gz",
1223
+ "image": "ACC_g245h/ACC_g245h-data.nii.gz"
1224
+ },
1225
+ {
1226
+ "label": "ACC_g136h/ACC_g136h-label.nii.gz",
1227
+ "image": "ACC_g136h/ACC_g136h-data.nii.gz"
1228
+ },
1229
+ {
1230
+ "label": "ACC_g452h/ACC_g452h-label.nii.gz",
1231
+ "image": "ACC_g452h/ACC_g452h-data.nii.gz"
1232
+ },
1233
+ {
1234
+ "label": "ACC_g455h/ACC_g455h-label.nii.gz",
1235
+ "image": "ACC_g455h/ACC_g455h-data.nii.gz"
1236
+ },
1237
+ {
1238
+ "label": "ACC_g650h/ACC_g650h-label.nii.gz",
1239
+ "image": "ACC_g650h/ACC_g650h-data.nii.gz"
1240
+ },
1241
+ {
1242
+ "label": "ACC_g165h/ACC_g165h-label.nii.gz",
1243
+ "image": "ACC_g165h/ACC_g165h-data.nii.gz"
1244
+ },
1245
+ {
1246
+ "label": "ACC_g485h/ACC_g485h-label.nii.gz",
1247
+ "image": "ACC_g485h/ACC_g485h-data.nii.gz"
1248
+ },
1249
+ {
1250
+ "label": "acc_g061h/acc_g061h-label.nii.gz",
1251
+ "image": "acc_g061h/acc_g061h-data.nii.gz"
1252
+ },
1253
+ {
1254
+ "label": "ACC_g445h/ACC_g445h-label.nii.gz",
1255
+ "image": "ACC_g445h/ACC_g445h-data.nii.gz"
1256
+ },
1257
+ {
1258
+ "label": "ACC_g486h/ACC_g486h-label.nii.gz",
1259
+ "image": "ACC_g486h/ACC_g486h-data.nii.gz"
1260
+ },
1261
+ {
1262
+ "label": "ACC_g341h/ACC_g341h-label.nii.gz",
1263
+ "image": "ACC_g341h/ACC_g341h-data.nii.gz"
1264
+ },
1265
+ {
1266
+ "label": "ACC_g354h/ACC_g354h-label.nii.gz",
1267
+ "image": "ACC_g354h/ACC_g354h-data.nii.gz"
1268
+ },
1269
+ {
1270
+ "label": "ACC_g672h/ACC_g672h-label.nii.gz",
1271
+ "image": "ACC_g672h/ACC_g672h-data.nii.gz"
1272
+ },
1273
+ {
1274
+ "label": "ACC_g553h/ACC_g553h-label.nii.gz",
1275
+ "image": "ACC_g553h/ACC_g553h-data.nii.gz"
1276
+ },
1277
+ {
1278
+ "label": "ACC_g656h/ACC_g656h-label.nii.gz",
1279
+ "image": "ACC_g656h/ACC_g656h-data.nii.gz"
1280
+ },
1281
+ {
1282
+ "label": "ACC_g609h/ACC_g609h-label.nii.gz",
1283
+ "image": "ACC_g609h/ACC_g609h-data.nii.gz"
1284
+ },
1285
+ {
1286
+ "label": "ACC_g333h/ACC_g333h-label.nii.gz",
1287
+ "image": "ACC_g333h/ACC_g333h-data.nii.gz"
1288
+ },
1289
+ {
1290
+ "label": "ACC_g284h/ACC_g284h-label.nii.gz",
1291
+ "image": "ACC_g284h/ACC_g284h-data.nii.gz"
1292
+ },
1293
+ {
1294
+ "label": "ACC_g738h/ACC_g738h-label.nii.gz",
1295
+ "image": "ACC_g738h/ACC_g738h-data.nii.gz"
1296
+ },
1297
+ {
1298
+ "label": "ACC_g651h/ACC_g651h-label.nii.gz",
1299
+ "image": "ACC_g651h/ACC_g651h-data.nii.gz"
1300
+ },
1301
+ {
1302
+ "label": "ACC_g663h/ACC_g663h-label.nii.gz",
1303
+ "image": "ACC_g663h/ACC_g663h-data.nii.gz"
1304
+ },
1305
+ {
1306
+ "label": "acc_g066h/acc_g066h-label.nii.gz",
1307
+ "image": "acc_g066h/acc_g066h-data.nii.gz"
1308
+ },
1309
+ {
1310
+ "label": "acc_g019/acc_g019-label.nii.gz",
1311
+ "image": "acc_g019/acc_g019-data.nii.gz"
1312
+ },
1313
+ {
1314
+ "label": "ACC_g276h/ACC_g276h-label.nii.gz",
1315
+ "image": "ACC_g276h/ACC_g276h-data.nii.gz"
1316
+ },
1317
+ {
1318
+ "label": "ACC_g435h/ACC_g435h-label.nii.gz",
1319
+ "image": "ACC_g435h/ACC_g435h-data.nii.gz"
1320
+ },
1321
+ {
1322
+ "label": "ACC_g212h/ACC_g212h-label.nii.gz",
1323
+ "image": "ACC_g212h/ACC_g212h-data.nii.gz"
1324
+ },
1325
+ {
1326
+ "label": "ACC_g157h/ACC_g157h-label.nii.gz",
1327
+ "image": "ACC_g157h/ACC_g157h-data.nii.gz"
1328
+ },
1329
+ {
1330
+ "label": "acc_g017/acc_g017-label.nii.gz",
1331
+ "image": "acc_g017/acc_g017-data.nii.gz"
1332
+ },
1333
+ {
1334
+ "label": "ACC_g362h/ACC_g362h-label.nii.gz",
1335
+ "image": "ACC_g362h/ACC_g362h-data.nii.gz"
1336
+ },
1337
+ {
1338
+ "label": "ACC_g113h/ACC_g113h-label.nii.gz",
1339
+ "image": "ACC_g113h/ACC_g113h-data.nii.gz"
1340
+ },
1341
+ {
1342
+ "label": "acc_g030/acc_g030-label.nii.gz",
1343
+ "image": "acc_g030/acc_g030-data.nii.gz"
1344
+ },
1345
+ {
1346
+ "label": "acc_g018/acc_g018-label.nii.gz",
1347
+ "image": "acc_g018/acc_g018-data.nii.gz"
1348
+ },
1349
+ {
1350
+ "label": "ACC_g359h/ACC_g359h-label.nii.gz",
1351
+ "image": "ACC_g359h/ACC_g359h-data.nii.gz"
1352
+ },
1353
+ {
1354
+ "label": "acc_g056h/acc_g056h-label.nii.gz",
1355
+ "image": "acc_g056h/acc_g056h-data.nii.gz"
1356
+ },
1357
+ {
1358
+ "label": "ACC_g692h/ACC_g692h-label.nii.gz",
1359
+ "image": "ACC_g692h/ACC_g692h-data.nii.gz"
1360
+ },
1361
+ {
1362
+ "label": "ACC_g125h/ACC_g125h-label.nii.gz",
1363
+ "image": "ACC_g125h/ACC_g125h-data.nii.gz"
1364
+ },
1365
+ {
1366
+ "label": "acc_g013/acc_g013-label.nii.gz",
1367
+ "image": "acc_g013/acc_g013-data.nii.gz"
1368
+ },
1369
+ {
1370
+ "label": "ACC_g305/ACC_g305-label.nii.gz",
1371
+ "image": "ACC_g305/ACC_g305-data.nii.gz"
1372
+ },
1373
+ {
1374
+ "label": "ACC_g569h/ACC_g569h-label.nii.gz",
1375
+ "image": "ACC_g569h/ACC_g569h-data.nii.gz"
1376
+ },
1377
+ {
1378
+ "label": "ACC_g642h/ACC_g642h-label.nii.gz",
1379
+ "image": "ACC_g642h/ACC_g642h-data.nii.gz"
1380
+ },
1381
+ {
1382
+ "label": "acc_g004/acc_g004-label.nii.gz",
1383
+ "image": "acc_g004/acc_g004-data.nii.gz"
1384
+ },
1385
+ {
1386
+ "label": "ACC_g713h/ACC_g713h-label.nii.gz",
1387
+ "image": "ACC_g713h/ACC_g713h-data.nii.gz"
1388
+ },
1389
+ {
1390
+ "label": "ACC_g358h/ACC_g358h-label.nii.gz",
1391
+ "image": "ACC_g358h/ACC_g358h-data.nii.gz"
1392
+ },
1393
+ {
1394
+ "label": "ACC_g368h/ACC_g368h-label.nii.gz",
1395
+ "image": "ACC_g368h/ACC_g368h-data.nii.gz"
1396
+ },
1397
+ {
1398
+ "label": "ACC_g725h/ACC_g725h-label.nii.gz",
1399
+ "image": "ACC_g725h/ACC_g725h-data.nii.gz"
1400
+ },
1401
+ {
1402
+ "label": "ACC_g110h/ACC_g110h-label.nii.gz",
1403
+ "image": "ACC_g110h/ACC_g110h-data.nii.gz"
1404
+ },
1405
+ {
1406
+ "label": "ACC_g533h/ACC_g533h-label.nii.gz",
1407
+ "image": "ACC_g533h/ACC_g533h-data.nii.gz"
1408
+ },
1409
+ {
1410
+ "label": "ACC_g275h/ACC_g275h-label.nii.gz",
1411
+ "image": "ACC_g275h/ACC_g275h-data.nii.gz"
1412
+ },
1413
+ {
1414
+ "label": "acc_g037/acc_g037-label.nii.gz",
1415
+ "image": "acc_g037/acc_g037-data.nii.gz"
1416
+ },
1417
+ {
1418
+ "label": "ACC_g749h/ACC_g749h-label.nii.gz",
1419
+ "image": "ACC_g749h/ACC_g749h-data.nii.gz"
1420
+ },
1421
+ {
1422
+ "label": "acc_g024/acc_g024-label.nii.gz",
1423
+ "image": "acc_g024/acc_g024-data.nii.gz"
1424
+ },
1425
+ {
1426
+ "label": "ACC_g705h/ACC_g705h-label.nii.gz",
1427
+ "image": "ACC_g705h/ACC_g705h-data.nii.gz"
1428
+ },
1429
+ {
1430
+ "label": "ACC_g106h/ACC_g106h-label.nii.gz",
1431
+ "image": "ACC_g106h/ACC_g106h-data.nii.gz"
1432
+ },
1433
+ {
1434
+ "label": "ACC_g668h/ACC_g668h-label.nii.gz",
1435
+ "image": "ACC_g668h/ACC_g668h-data.nii.gz"
1436
+ },
1437
+ {
1438
+ "label": "ACC_g481h/ACC_g481h-label.nii.gz",
1439
+ "image": "ACC_g481h/ACC_g481h-data.nii.gz"
1440
+ },
1441
+ {
1442
+ "label": "ACC_g653h/ACC_g653h-label.nii.gz",
1443
+ "image": "ACC_g653h/ACC_g653h-data.nii.gz"
1444
+ }
1445
+ ],
1446
+ "testing": [
1447
+ {
1448
+ "label": "ACC_g515h/ACC_g515h-label.nii.gz",
1449
+ "image": "ACC_g515h/ACC_g515h-data.nii.gz"
1450
+ },
1451
+ {
1452
+ "label": "ACC_g123h/ACC_g123h-label.nii.gz",
1453
+ "image": "ACC_g123h/ACC_g123h-data.nii.gz"
1454
+ },
1455
+ {
1456
+ "label": "ACC_g747h/ACC_g747h-label.nii.gz",
1457
+ "image": "ACC_g747h/ACC_g747h-data.nii.gz"
1458
+ },
1459
+ {
1460
+ "label": "ACC_g131h/ACC_g131h-label.nii.gz",
1461
+ "image": "ACC_g131h/ACC_g131h-data.nii.gz"
1462
+ },
1463
+ {
1464
+ "label": "ACC_g607h/ACC_g607h-label.nii.gz",
1465
+ "image": "ACC_g607h/ACC_g607h-data.nii.gz"
1466
+ },
1467
+ {
1468
+ "label": "acc_g065h/acc_g065h-label.nii.gz",
1469
+ "image": "acc_g065h/acc_g065h-data.nii.gz"
1470
+ },
1471
+ {
1472
+ "label": "ACC_g740h/ACC_g740h-label.nii.gz",
1473
+ "image": "ACC_g740h/ACC_g740h-data.nii.gz"
1474
+ },
1475
+ {
1476
+ "label": "acc_g003/acc_g003-label.nii.gz",
1477
+ "image": "acc_g003/acc_g003-data.nii.gz"
1478
+ },
1479
+ {
1480
+ "label": "acc_g040/acc_g040-label.nii.gz",
1481
+ "image": "acc_g040/acc_g040-data.nii.gz"
1482
+ },
1483
+ {
1484
+ "label": "ACC_g177h/ACC_g177h-label.nii.gz",
1485
+ "image": "ACC_g177h/ACC_g177h-data.nii.gz"
1486
+ },
1487
+ {
1488
+ "label": "ACC_g386h/ACC_g386h-label.nii.gz",
1489
+ "image": "ACC_g386h/ACC_g386h-data.nii.gz"
1490
+ },
1491
+ {
1492
+ "label": "ACC_g145h/ACC_g145h-label.nii.gz",
1493
+ "image": "ACC_g145h/ACC_g145h-data.nii.gz"
1494
+ },
1495
+ {
1496
+ "label": "ACC_g398h/ACC_g398h-label.nii.gz",
1497
+ "image": "ACC_g398h/ACC_g398h-data.nii.gz"
1498
+ },
1499
+ {
1500
+ "label": "ACC_g351h/ACC_g351h-label.nii.gz",
1501
+ "image": "ACC_g351h/ACC_g351h-data.nii.gz"
1502
+ },
1503
+ {
1504
+ "label": "ACC_g529h/ACC_g529h-label.nii.gz",
1505
+ "image": "ACC_g529h/ACC_g529h-data.nii.gz"
1506
+ },
1507
+ {
1508
+ "label": "acc_g016/acc_g016-label.nii.gz",
1509
+ "image": "acc_g016/acc_g016-data.nii.gz"
1510
+ },
1511
+ {
1512
+ "label": "ACC_g633h/ACC_g633h-label.nii.gz",
1513
+ "image": "ACC_g633h/ACC_g633h-data.nii.gz"
1514
+ },
1515
+ {
1516
+ "label": "ACC_g576h/ACC_g576h-label.nii.gz",
1517
+ "image": "ACC_g576h/ACC_g576h-data.nii.gz"
1518
+ },
1519
+ {
1520
+ "label": "ACC_g516h/ACC_g516h-label.nii.gz",
1521
+ "image": "ACC_g516h/ACC_g516h-data.nii.gz"
1522
+ },
1523
+ {
1524
+ "label": "ACC_g523h/ACC_g523h-label.nii.gz",
1525
+ "image": "ACC_g523h/ACC_g523h-data.nii.gz"
1526
+ },
1527
+ {
1528
+ "label": "ACC_g150h/ACC_g150h-label.nii.gz",
1529
+ "image": "ACC_g150h/ACC_g150h-data.nii.gz"
1530
+ },
1531
+ {
1532
+ "label": "ACC_g564h/ACC_g564h-label.nii.gz",
1533
+ "image": "ACC_g564h/ACC_g564h-data.nii.gz"
1534
+ },
1535
+ {
1536
+ "label": "ACC_g191h/ACC_g191h-label.nii.gz",
1537
+ "image": "ACC_g191h/ACC_g191h-data.nii.gz"
1538
+ },
1539
+ {
1540
+ "label": "ACC_g080h/ACC_g080h-label.nii.gz",
1541
+ "image": "ACC_g080h/ACC_g080h-data.nii.gz"
1542
+ },
1543
+ {
1544
+ "label": "ACC_g587h/ACC_g587h-label.nii.gz",
1545
+ "image": "ACC_g587h/ACC_g587h-data.nii.gz"
1546
+ },
1547
+ {
1548
+ "label": "ACC_g458h/ACC_g458h-label.nii.gz",
1549
+ "image": "ACC_g458h/ACC_g458h-data.nii.gz"
1550
+ },
1551
+ {
1552
+ "label": "ACC_g223h/ACC_g223h-label.nii.gz",
1553
+ "image": "ACC_g223h/ACC_g223h-data.nii.gz"
1554
+ },
1555
+ {
1556
+ "label": "acc_g032/acc_g032-label.nii.gz",
1557
+ "image": "acc_g032/acc_g032-data.nii.gz"
1558
+ },
1559
+ {
1560
+ "label": "ACC_g176h/ACC_g176h-label.nii.gz",
1561
+ "image": "ACC_g176h/ACC_g176h-data.nii.gz"
1562
+ },
1563
+ {
1564
+ "label": "acc_g007/acc_g007-label.nii.gz",
1565
+ "image": "acc_g007/acc_g007-data.nii.gz"
1566
+ },
1567
+ {
1568
+ "label": "ACC_g675h/ACC_g675h-label.nii.gz",
1569
+ "image": "ACC_g675h/ACC_g675h-data.nii.gz"
1570
+ },
1571
+ {
1572
+ "label": "ACC_g669h/ACC_g669h-label.nii.gz",
1573
+ "image": "ACC_g669h/ACC_g669h-data.nii.gz"
1574
+ },
1575
+ {
1576
+ "label": "ACC_g710h/ACC_g710h-label.nii.gz",
1577
+ "image": "ACC_g710h/ACC_g710h-data.nii.gz"
1578
+ },
1579
+ {
1580
+ "label": "ACC_g304h/ACC_g304h-label.nii.gz",
1581
+ "image": "ACC_g304h/ACC_g304h-data.nii.gz"
1582
+ },
1583
+ {
1584
+ "label": "ACC_g423h/ACC_g423h-label.nii.gz",
1585
+ "image": "ACC_g423h/ACC_g423h-data.nii.gz"
1586
+ },
1587
+ {
1588
+ "label": "acc_g048h/acc_g048h-label.nii.gz",
1589
+ "image": "acc_g048h/acc_g048h-data.nii.gz"
1590
+ },
1591
+ {
1592
+ "label": "acc_g076h/acc_g076h-label.nii.gz",
1593
+ "image": "acc_g076h/acc_g076h-data.nii.gz"
1594
+ },
1595
+ {
1596
+ "label": "ACC_g294h/ACC_g294h-label.nii.gz",
1597
+ "image": "ACC_g294h/ACC_g294h-data.nii.gz"
1598
+ },
1599
+ {
1600
+ "label": "ACC_g520h/ACC_g520h-label.nii.gz",
1601
+ "image": "ACC_g520h/ACC_g520h-data.nii.gz"
1602
+ },
1603
+ {
1604
+ "label": "ACC_g372h/ACC_g372h-label.nii.gz",
1605
+ "image": "ACC_g372h/ACC_g372h-data.nii.gz"
1606
+ },
1607
+ {
1608
+ "label": "ACC_g641h/ACC_g641h-label.nii.gz",
1609
+ "image": "ACC_g641h/ACC_g641h-data.nii.gz"
1610
+ },
1611
+ {
1612
+ "label": "ACC_g216h/ACC_g216h-label.nii.gz",
1613
+ "image": "ACC_g216h/ACC_g216h-data.nii.gz"
1614
+ },
1615
+ {
1616
+ "label": "ACC_g573h/ACC_g573h-label.nii.gz",
1617
+ "image": "ACC_g573h/ACC_g573h-data.nii.gz"
1618
+ },
1619
+ {
1620
+ "label": "ACC_g396h/ACC_g396h-label.nii.gz",
1621
+ "image": "ACC_g396h/ACC_g396h-data.nii.gz"
1622
+ },
1623
+ {
1624
+ "label": "ACC_g143h/ACC_g143h-label.nii.gz",
1625
+ "image": "ACC_g143h/ACC_g143h-data.nii.gz"
1626
+ },
1627
+ {
1628
+ "label": "ACC_g229h/ACC_g229h-label.nii.gz",
1629
+ "image": "ACC_g229h/ACC_g229h-data.nii.gz"
1630
+ },
1631
+ {
1632
+ "label": "ACC_g250h/ACC_g250h-label.nii.gz",
1633
+ "image": "ACC_g250h/ACC_g250h-data.nii.gz"
1634
+ },
1635
+ {
1636
+ "label": "acc_g023/acc_g023-label.nii.gz",
1637
+ "image": "acc_g023/acc_g023-data.nii.gz"
1638
+ },
1639
+ {
1640
+ "label": "acc_g074h/acc_g074h-label.nii.gz",
1641
+ "image": "acc_g074h/acc_g074h-data.nii.gz"
1642
+ },
1643
+ {
1644
+ "label": "ACC_g174h/ACC_g174h-label.nii.gz",
1645
+ "image": "ACC_g174h/ACC_g174h-data.nii.gz"
1646
+ },
1647
+ {
1648
+ "label": "ACC_g107h/ACC_g107h-label.nii.gz",
1649
+ "image": "ACC_g107h/ACC_g107h-data.nii.gz"
1650
+ },
1651
+ {
1652
+ "label": "acc_g087h/acc_g087h-label.nii.gz",
1653
+ "image": "acc_g087h/acc_g087h-data.nii.gz"
1654
+ },
1655
+ {
1656
+ "label": "ACC_g117h/ACC_g117h-label.nii.gz",
1657
+ "image": "ACC_g117h/ACC_g117h-data.nii.gz"
1658
+ },
1659
+ {
1660
+ "label": "ACC_g267h/ACC_g267h-label.nii.gz",
1661
+ "image": "ACC_g267h/ACC_g267h-data.nii.gz"
1662
+ },
1663
+ {
1664
+ "label": "ACC_g361h/ACC_g361h-label.nii.gz",
1665
+ "image": "ACC_g361h/ACC_g361h-data.nii.gz"
1666
+ },
1667
+ {
1668
+ "label": "ACC_g683h/ACC_g683h-label.nii.gz",
1669
+ "image": "ACC_g683h/ACC_g683h-data.nii.gz"
1670
+ },
1671
+ {
1672
+ "label": "ACC_g644h/ACC_g644h-label.nii.gz",
1673
+ "image": "ACC_g644h/ACC_g644h-data.nii.gz"
1674
+ },
1675
+ {
1676
+ "label": "ACC_g479h/ACC_g479h-label.nii.gz",
1677
+ "image": "ACC_g479h/ACC_g479h-data.nii.gz"
1678
+ },
1679
+ {
1680
+ "label": "ACC_g156h/ACC_g156h-label.nii.gz",
1681
+ "image": "ACC_g156h/ACC_g156h-data.nii.gz"
1682
+ },
1683
+ {
1684
+ "label": "ACC_g467h/ACC_g467h-label.nii.gz",
1685
+ "image": "ACC_g467h/ACC_g467h-data.nii.gz"
1686
+ }
1687
+ ]
1688
+ }
configs/evaluate.yaml ADDED
@@ -0,0 +1,65 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ validate#postprocessing:
3
+ _target_: Compose
4
+ transforms:
5
+ - _target_: Activationsd
6
+ keys: pred
7
+ softmax: true
8
+ - _target_: Invertd
9
+ keys:
10
+ - pred
11
+ - label
12
+ transform: "@validate#preprocessing"
13
+ orig_keys: image
14
+ meta_key_postfix: meta_dict
15
+ nearest_interp:
16
+ - false
17
+ - true
18
+ to_tensor: true
19
+ - _target_: AsDiscreted
20
+ keys:
21
+ - pred
22
+ - label
23
+ argmax:
24
+ - true
25
+ - false
26
+ to_onehot: 8
27
+ - _target_: CopyItemsd
28
+ keys: "pred"
29
+ times: 1
30
+ names: "pred_save"
31
+ - _target_: AsDiscreted
32
+ keys:
33
+ - pred_save
34
+ argmax:
35
+ - true
36
+ - _target_: SaveImaged
37
+ keys: pred_save
38
+ meta_keys: pred_meta_dict
39
+ output_dir: "@output_dir"
40
+ resample: false
41
+ squeeze_end_dims: true
42
+ validate#dataset:
43
+ _target_: Dataset
44
+ data: "@val_datalist"
45
+ transform: "@validate#preprocessing"
46
+ validate#handlers:
47
+ - _target_: CheckpointLoader
48
+ load_path: "$@ckpt_dir + '/model.pt'"
49
+ load_dict:
50
+ model: "@network"
51
+ - _target_: StatsHandler
52
+ iteration_log: false
53
+ - _target_: MetricsSaver
54
+ save_dir: "@output_dir"
55
+ metrics:
56
+ - val_mean_dice
57
+ - val_acc
58
+ metric_details:
59
+ - val_mean_dice
60
+ batch_transform: "$monai.handlers.from_engine(['image_meta_dict'])"
61
+ summary_ops: "*"
62
+ initialize:
63
+ - "$setattr(torch.backends.cudnn, 'benchmark', True)"
64
+ run:
65
+ - "$@validate#evaluator.run()"
configs/inference.yaml ADDED
@@ -0,0 +1,117 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ imports:
3
+ - "$import glob"
4
+ - "$import os"
5
+ input_channels: 1
6
+ output_classes: 8
7
+ arch_ckpt_path: "$@bundle_root + '/models/search_code_18590.pt'"
8
+ arch_ckpt: "$torch.load(@arch_ckpt_path, map_location=torch.device('cuda'))"
9
+ bundle_root: "."
10
+ output_dir: "$@bundle_root + '/eval'"
11
+ dataset_dir: "/data/"
12
+ data_list_file_path: "$@bundle_root + '/configs/dataset_0.json'"
13
+ datalist: "$monai.data.load_decathlon_datalist(@data_list_file_path, data_list_key='testing',
14
+ base_dir=@dataset_dir)"
15
+ device: "$torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')"
16
+ dints_space:
17
+ _target_: monai.networks.nets.TopologyInstance
18
+ channel_mul: 1
19
+ num_blocks: 12
20
+ num_depths: 4
21
+ use_downsample: true
22
+ arch_code:
23
+ - "$@arch_ckpt['arch_code_a']"
24
+ - "$@arch_ckpt['arch_code_c']"
25
+ device: "$torch.device('cuda')"
26
+ network_def:
27
+ _target_: monai.networks.nets.DiNTS
28
+ dints_space: "@dints_space"
29
+ in_channels: "@input_channels"
30
+ num_classes: "@output_classes"
31
+ use_downsample: true
32
+ node_a: "$torch.from_numpy(@arch_ckpt['node_a'])"
33
+ network: "$@network_def.to(@device)"
34
+ preprocessing:
35
+ _target_: Compose
36
+ transforms:
37
+ - _target_: LoadImaged
38
+ keys: image
39
+ - _target_: EnsureChannelFirstd
40
+ keys: image
41
+ - _target_: Orientationd
42
+ keys: image
43
+ axcodes: RAS
44
+ - _target_: Spacingd
45
+ keys: image
46
+ pixdim:
47
+ - 1
48
+ - 1
49
+ - 1
50
+ mode: bilinear
51
+ - _target_: ScaleIntensityRanged
52
+ keys: image
53
+ a_min: -500
54
+ a_max: 500
55
+ b_min: 0
56
+ b_max: 1
57
+ clip: true
58
+ - _target_: EnsureTyped
59
+ keys: image
60
+ dataset:
61
+ _target_: Dataset
62
+ data: "@datalist"
63
+ transform: "@preprocessing"
64
+ dataloader:
65
+ _target_: DataLoader
66
+ dataset: "@dataset"
67
+ batch_size: 1
68
+ shuffle: false
69
+ num_workers: 4
70
+ inferer:
71
+ _target_: SlidingWindowInferer
72
+ roi_size:
73
+ - 96
74
+ - 96
75
+ - 96
76
+ sw_batch_size: 4
77
+ overlap: 0.625
78
+ postprocessing:
79
+ _target_: Compose
80
+ transforms:
81
+ - _target_: Activationsd
82
+ keys: pred
83
+ softmax: true
84
+ - _target_: Invertd
85
+ keys: pred
86
+ transform: "@preprocessing"
87
+ orig_keys: image
88
+ meta_key_postfix: meta_dict
89
+ nearest_interp: false
90
+ to_tensor: true
91
+ - _target_: AsDiscreted
92
+ keys: pred
93
+ argmax: true
94
+ - _target_: SaveImaged
95
+ keys: pred
96
+ meta_keys: pred_meta_dict
97
+ output_dir: "@output_dir"
98
+ handlers:
99
+ - _target_: CheckpointLoader
100
+ load_path: "$@bundle_root + '/models/model.pt'"
101
+ load_dict:
102
+ model: "@network"
103
+ - _target_: StatsHandler
104
+ iteration_log: false
105
+ evaluator:
106
+ _target_: SupervisedEvaluator
107
+ device: "@device"
108
+ val_data_loader: "@dataloader"
109
+ network: "@network"
110
+ inferer: "@inferer"
111
+ postprocessing: "@postprocessing"
112
+ val_handlers: "@handlers"
113
+ amp: true
114
+ initialize:
115
+ - "$setattr(torch.backends.cudnn, 'benchmark', True)"
116
+ run:
117
configs/inference_trt.yaml ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ imports:
3
+ - "$import glob"
4
+ - "$import os"
5
+ - "$import torch_tensorrt"
6
+ handlers#0#_disabled_: true
7
+ network_def: "$torch.jit.load(@bundle_root + '/models/model.ts')"
8
+ evaluator#amp: false
configs/logging.conf ADDED
@@ -0,0 +1,21 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ [loggers]
2
+ keys=root
3
+
4
+ [handlers]
5
+ keys=consoleHandler
6
+
7
+ [formatters]
8
+ keys=fullFormatter
9
+
10
+ [logger_root]
11
+ level=INFO
12
+ handlers=consoleHandler
13
+
14
+ [handler_consoleHandler]
15
+ class=StreamHandler
16
+ level=INFO
17
+ formatter=fullFormatter
18
+ args=(sys.stdout,)
19
+
20
+ [formatter_fullFormatter]
21
+ format=%(asctime)s - %(name)s - %(levelname)s - %(message)s
configs/metadata.json ADDED
@@ -0,0 +1,97 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "schema": "https://github.com/Project-MONAI/MONAI-extra-test-data/releases/download/0.8.1/meta_schema_20240725.json",
3
+ "version": "0.0.5",
4
+ "changelog": {
5
+ "0.0.5": "update to huggingface hosting",
6
+ "0.0.4": "Set image_only to False",
7
+ "0.0.3": "Update for stable MONAI version",
8
+ "0.0.2": "Retrain with new MONAI",
9
+ "0.0.1": "initialize the model package structure"
10
+ },
11
+ "monai_version": "1.3.0",
12
+ "pytorch_version": "1.13.1",
13
+ "numpy_version": "1.22.2",
14
+ "required_packages_version": {
15
+ "fire": "0.5.0",
16
+ "nibabel": "5.1.0",
17
+ "pytorch-ignite": "0.4.11",
18
+ "pyyaml": "6.0.2"
19
+ },
20
+ "supported_apps": {},
21
+ "name": "Abdominal multi-organ segmentation",
22
+ "task": "Multi-organ segmentation in abdominal CT",
23
+ "description": "DiNTS architectures for volumetric (3D) segmentation of the abdominal from CT image",
24
+ "authors": "Chen Shen, Holger R. Roth, Kazunari Misawa, Kensaku Mori",
25
+ "copyright": "",
26
+ "data_source": "Aichi Cancer Center, Japan",
27
+ "data_type": "nibabel",
28
+ "image_classes": "single channel data, intensity scaled to [0, 1]",
29
+ "label_classes": "eight channels data, 1 is artery, 2 is portal vein, 3 is liver, 4 is spleen, 5 is stomach, 6 is gallbladder, 7 is pancreas, 0 is everything else",
30
+ "pred_classes": "8 channels OneHot data, 1 is artery, 2 is portal vein, 3 is liver, 4 is spleen, 5 is stomach, 6 is gallbladder, 7 is pancreas, 0 is background",
31
+ "eval_metrics": {
32
+ "mean_dice": 0.88
33
+ },
34
+ "references": [
35
+ "He, Y., Yang, D., Roth, H., Zhao, C. and Xu, D., 2021. Dints: Differentiable neural network topology search for 3d medical image segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 5841-5850).",
36
+ "Roth, H., Shen C, Oda H., Sugino T., Oda M., Hayashi Y., Misawa K., Mori K., 2018. A multi-scale pyramid of 3D fully convolutional networks for abdominal multi-organ segmentation. International conference on medical image computing and computer-assisted intervention",
37
+ "Shen, C., Roth, H. R., Nath, V., Hayashi, Y., Oda, M., Misawa, K., Mori, K., 2022. Effective hyperparameter optimization with proxy data for multi-organ segmentation. In Medical Imaging 2022: Image Processing (Vol. 12032, pp. 200-206)"
38
+ ],
39
+ "network_data_format": {
40
+ "inputs": {
41
+ "image": {
42
+ "type": "image",
43
+ "format": "hounsfield",
44
+ "modality": "CT",
45
+ "num_channels": 1,
46
+ "spatial_shape": [
47
+ 96,
48
+ 96,
49
+ 96
50
+ ],
51
+ "dtype": "float32",
52
+ "value_range": [
53
+ 0,
54
+ 1
55
+ ],
56
+ "is_patch_data": true,
57
+ "channel_def": {
58
+ "0": "image"
59
+ }
60
+ }
61
+ },
62
+ "outputs": {
63
+ "pred": {
64
+ "type": "image",
65
+ "format": "segmentation",
66
+ "num_channels": 8,
67
+ "spatial_shape": [
68
+ 96,
69
+ 96,
70
+ 96
71
+ ],
72
+ "dtype": "float32",
73
+ "value_range": [
74
+ 0,
75
+ 1,
76
+ 2,
77
+ 3,
78
+ 4,
79
+ 5,
80
+ 6,
81
+ 7
82
+ ],
83
+ "is_patch_data": true,
84
+ "channel_def": {
85
+ "0": "background",
86
+ "1": "artery",
87
+ "2": "portal vein",
88
+ "3": "liver",
89
+ "4": "spleen",
90
+ "5": "stomach",
91
+ "6": "gallbladder",
92
+ "7": "pancreas"
93
+ }
94
+ }
95
+ }
96
+ }
97
+ }
configs/multi_gpu_train.yaml ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ device: "$torch.device(f'cuda:{dist.get_rank()}')"
3
+ network:
4
+ _target_: torch.nn.parallel.DistributedDataParallel
5
+ module: "$@network_def.to(@device)"
6
+ find_unused_parameters: true
7
+ device_ids:
8
+ - "@device"
9
+ optimizer#lr: "$0.025*dist.get_world_size()"
10
+ lr_scheduler#step_size: "$80*dist.get_world_size()"
11
+ train#handlers:
12
+ - _target_: LrScheduleHandler
13
+ lr_scheduler: "@lr_scheduler"
14
+ print_lr: true
15
+ - _target_: ValidationHandler
16
+ validator: "@validate#evaluator"
17
+ epoch_level: true
18
+ interval: "$10*dist.get_world_size()"
19
+ - _target_: StatsHandler
20
+ tag_name: train_loss
21
+ output_transform: "$monai.handlers.from_engine(['loss'], first=True)"
22
+ - _target_: TensorBoardStatsHandler
23
+ log_dir: "@output_dir"
24
+ tag_name: train_loss
25
+ output_transform: "$monai.handlers.from_engine(['loss'], first=True)"
26
+ train#trainer#max_epochs: "$400*dist.get_world_size()"
27
+ train#trainer#train_handlers: "$@train#handlers[: -2 if dist.get_rank() > 0 else None]"
28
+ validate#evaluator#val_handlers: "$None if dist.get_rank() > 0 else @validate#handlers"
29
+ initialize:
30
+ - "$import torch.distributed as dist"
31
+ - "$dist.init_process_group(backend='nccl')"
32
+ - "$torch.cuda.set_device(@device)"
33
+ - "$monai.utils.set_determinism(seed=123)"
34
+ - "$setattr(torch.backends.cudnn, 'benchmark', True)"
35
+ run:
36
+ - "$@train#trainer.run()"
37
+ finalize:
38
+ - $dist.is_initialized() and dist.destroy_process_group()
39
+ train_data_partition: "$monai.data.partition_dataset(data=@train_datalist, num_partitions=dist.get_world_size(),
40
+ shuffle=True, even_divisible=True,)[dist.get_rank()]"
41
+ train#dataset:
42
+ _target_: CacheDataset
43
+ data: "@train_data_partition"
44
+ transform: "@train#preprocessing"
45
+ cache_rate: 0.2
46
+ num_workers: 4
47
+ val_data_partition: "$monai.data.partition_dataset(data=@val_datalist, num_partitions=dist.get_world_size(),
48
+ shuffle=False, even_divisible=False,)[dist.get_rank()]"
49
+ validate#dataset:
50
+ _target_: CacheDataset
51
+ data: "@val_data_partition"
52
+ transform: "@validate#preprocessing"
53
+ cache_rate: 0.2
54
+ num_workers: 4
configs/search.yaml ADDED
@@ -0,0 +1,276 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ imports:
3
+ - "$from scipy import ndimage"
4
+ arch_ckpt_path: models
5
+ amp: true
6
+ data_file_base_dir: /workspace/data/msd/Task07_Pancreas
7
+ data_list_file_path: configs/dataset_0.json
8
+ determ: true
9
+ input_channels: 1
10
+ learning_rate: 0.025
11
+ learning_rate_arch: 0.001
12
+ learning_rate_milestones:
13
+ - 0.4
14
+ - 0.8
15
+ num_images_per_batch: 1
16
+ num_epochs: 1430
17
+ num_epochs_per_validation: 100
18
+ num_epochs_warmup: 715
19
+ num_patches_per_image: 1
20
+ num_sw_batch_size: 6
21
+ output_classes: 3
22
+ overlap_ratio: 0.625
23
+ patch_size:
24
+ - 96
25
+ - 96
26
+ - 96
27
+ patch_size_valid:
28
+ - 96
29
+ - 96
30
+ - 96
31
+ ram_cost_factor: 0.8
32
+ image_key: image
33
+ label_key: label
34
+ transform_train:
35
+ _target_: Compose
36
+ transforms:
37
+ - _target_: LoadImaged
38
+ keys:
39
+ - "@image_key"
40
+ - "@label_key"
41
+ image_only: false
42
+ - _target_: EnsureChannelFirstd
43
+ keys:
44
+ - "@image_key"
45
+ - "@label_key"
46
+ - _target_: Orientationd
47
+ keys:
48
+ - "@image_key"
49
+ - "@label_key"
50
+ axcodes: RAS
51
+ - _target_: Spacingd
52
+ keys:
53
+ - "@image_key"
54
+ - "@label_key"
55
+ pixdim:
56
+ - 1
57
+ - 1
58
+ - 1
59
+ mode:
60
+ - bilinear
61
+ - nearest
62
+ align_corners:
63
+ - true
64
+ - true
65
+ - _target_: CastToTyped
66
+ keys: "@image_key"
67
+ dtype: "$torch.float32"
68
+ - _target_: ScaleIntensityRanged
69
+ keys: "@image_key"
70
+ a_min: -87
71
+ a_max: 199
72
+ b_min: 0
73
+ b_max: 1
74
+ clip: true
75
+ - _target_: CastToTyped
76
+ keys:
77
+ - "@image_key"
78
+ - "@label_key"
79
+ dtype:
80
+ - "$np.float16"
81
+ - "$np.uint8"
82
+ - _target_: CopyItemsd
83
+ keys: "@label_key"
84
+ times: 1
85
+ names:
86
+ - label4crop
87
+ - _target_: Lambdad
88
+ keys: label4crop
89
+ func: "$lambda x, s=@output_classes: np.concatenate(tuple([ndimage.binary_dilation((x==_k).astype(x.dtype), iterations=48).astype(float) for _k in range(s)]), axis=0)"
90
+ overwrite: true
91
+ - _target_: EnsureTyped
92
+ keys:
93
+ - "@image_key"
94
+ - "@label_key"
95
+ - _target_: CastToTyped
96
+ keys: "@image_key"
97
+ dtype: "$torch.float32"
98
+ - _target_: SpatialPadd
99
+ keys:
100
+ - "@image_key"
101
+ - "@label_key"
102
+ - label4crop
103
+ spatial_size: "@patch_size"
104
+ mode:
105
+ - reflect
106
+ - constant
107
+ - constant
108
+ - _target_: RandCropByLabelClassesd
109
+ keys:
110
+ - "@image_key"
111
+ - "@label_key"
112
+ label_key: label4crop
113
+ num_classes: "@output_classes"
114
+ ratios: "$[1,] * @output_classes"
115
+ spatial_size: "@patch_size"
116
+ num_samples: "@num_patches_per_image"
117
+ - _target_: Lambdad
118
+ keys: label4crop
119
+ func: "$lambda x: 0"
120
+ - _target_: RandRotated
121
+ keys:
122
+ - "@image_key"
123
+ - "@label_key"
124
+ range_x: 0.3
125
+ range_y: 0.3
126
+ range_z: 0.3
127
+ mode:
128
+ - bilinear
129
+ - nearest
130
+ prob: 0.2
131
+ - _target_: RandZoomd
132
+ keys:
133
+ - "@image_key"
134
+ - "@label_key"
135
+ min_zoom: 0.8
136
+ max_zoom: 1.2
137
+ mode:
138
+ - trilinear
139
+ - nearest
140
+ align_corners:
141
+ - null
142
+ - null
143
+ prob: 0.16
144
+ - _target_: RandGaussianSmoothd
145
+ keys: "@image_key"
146
+ sigma_x:
147
+ - 0.5
148
+ - 1.15
149
+ sigma_y:
150
+ - 0.5
151
+ - 1.15
152
+ sigma_z:
153
+ - 0.5
154
+ - 1.15
155
+ prob: 0.15
156
+ - _target_: RandScaleIntensityd
157
+ keys: "@image_key"
158
+ factors: 0.3
159
+ prob: 0.5
160
+ - _target_: RandShiftIntensityd
161
+ keys: "@image_key"
162
+ offsets: 0.1
163
+ prob: 0.5
164
+ - _target_: RandGaussianNoised
165
+ keys: "@image_key"
166
+ std: 0.01
167
+ prob: 0.15
168
+ - _target_: RandFlipd
169
+ keys:
170
+ - "@image_key"
171
+ - "@label_key"
172
+ spatial_axis: 0
173
+ prob: 0.5
174
+ - _target_: RandFlipd
175
+ keys:
176
+ - "@image_key"
177
+ - "@label_key"
178
+ spatial_axis: 1
179
+ prob: 0.5
180
+ - _target_: RandFlipd
181
+ keys:
182
+ - "@image_key"
183
+ - "@label_key"
184
+ spatial_axis: 2
185
+ prob: 0.5
186
+ - _target_: CastToTyped
187
+ keys:
188
+ - "@image_key"
189
+ - "@label_key"
190
+ dtype:
191
+ - "$torch.float32"
192
+ - "$torch.uint8"
193
+ - _target_: ToTensord
194
+ keys:
195
+ - "@image_key"
196
+ - "@label_key"
197
+ transform_validation:
198
+ _target_: Compose
199
+ transforms:
200
+ - _target_: LoadImaged
201
+ keys:
202
+ - "@image_key"
203
+ - "@label_key"
204
+ - _target_: EnsureChannelFirstd
205
+ keys:
206
+ - "@image_key"
207
+ - "@label_key"
208
+ - _target_: Orientationd
209
+ keys:
210
+ - "@image_key"
211
+ - "@label_key"
212
+ axcodes: RAS
213
+ - _target_: Spacingd
214
+ keys:
215
+ - "@image_key"
216
+ - "@label_key"
217
+ pixdim:
218
+ - 1
219
+ - 1
220
+ - 1
221
+ mode:
222
+ - bilinear
223
+ - nearest
224
+ align_corners:
225
+ - true
226
+ - true
227
+ - _target_: CastToTyped
228
+ keys: "@image_key"
229
+ dtype: "$torch.float32"
230
+ - _target_: ScaleIntensityRanged
231
+ keys: "@image_key"
232
+ a_min: -87
233
+ a_max: 199
234
+ b_min: 0
235
+ b_max: 1
236
+ clip: true
237
+ - _target_: CastToTyped
238
+ keys:
239
+ - "@image_key"
240
+ - "@label_key"
241
+ dtype:
242
+ - "$np.float16"
243
+ - "$np.uint8"
244
+ - _target_: CastToTyped
245
+ keys:
246
+ - "@image_key"
247
+ - "@label_key"
248
+ dtype:
249
+ - "$torch.float32"
250
+ - "$torch.uint8"
251
+ - _target_: ToTensord
252
+ keys:
253
+ - "@image_key"
254
+ - "@label_key"
255
+ loss:
256
+ _target_: DiceCELoss
257
+ include_background: false
258
+ to_onehot_y: true
259
+ softmax: true
260
+ squared_pred: true
261
+ batch: true
262
+ smooth_nr: 0.00001
263
+ smooth_dr: 0.00001
264
+ dints_space:
265
+ _target_: monai.networks.nets.TopologySearch
266
+ channel_mul: 0.5
267
+ num_blocks: 12
268
+ num_depths: 4
269
+ use_downsample: true
270
+ device: "$torch.device('cuda')"
271
+ network:
272
+ _target_: monai.networks.nets.DiNTS
273
+ dints_space: "@dints_space"
274
+ in_channels: "@input_channels"
275
+ num_classes: "@output_classes"
276
+ use_downsample: true
configs/train.yaml ADDED
@@ -0,0 +1,356 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ imports:
3
+ - "$import glob"
4
+ - "$import json"
5
+ - "$import os"
6
+ - "$import ignite"
7
+ - "$from scipy import ndimage"
8
+ input_channels: 1
9
+ output_classes: 8
10
+ arch_ckpt_path: "$@bundle_root + '/models/search_code_18590.pt'"
11
+ arch_ckpt: "$torch.load(@arch_ckpt_path, map_location=torch.device('cuda'))"
12
+ bundle_root: "."
13
+ ckpt_dir: "$@bundle_root + '/models'"
14
+ output_dir: "$@bundle_root + '/eval'"
15
+ dataset_dir: "/data/"
16
+ data_list_file_path: "$@bundle_root + '/configs/dataset_0.json'"
17
+ train_datalist: "$monai.data.load_decathlon_datalist(@data_list_file_path, data_list_key='training',
18
+ base_dir=@dataset_dir)"
19
+ val_datalist: "$monai.data.load_decathlon_datalist(@data_list_file_path, data_list_key='validation',
20
+ base_dir=@dataset_dir)"
21
+ device: "$torch.device('cuda:0' if torch.cuda.is_available() else 'cpu')"
22
+ dints_space:
23
+ _target_: monai.networks.nets.TopologyInstance
24
+ channel_mul: 1
25
+ num_blocks: 12
26
+ num_depths: 4
27
+ use_downsample: true
28
+ arch_code:
29
+ - "$@arch_ckpt['arch_code_a']"
30
+ - "$@arch_ckpt['arch_code_c']"
31
+ device: "$torch.device('cuda')"
32
+ network_def:
33
+ _target_: monai.networks.nets.DiNTS
34
+ dints_space: "@dints_space"
35
+ in_channels: "@input_channels"
36
+ num_classes: "@output_classes"
37
+ use_downsample: true
38
+ node_a: "$@arch_ckpt['node_a']"
39
+ network: "$@network_def.to(@device)"
40
+ loss:
41
+ _target_: DiceCELoss
42
+ include_background: false
43
+ to_onehot_y: true
44
+ softmax: true
45
+ squared_pred: true
46
+ batch: true
47
+ smooth_nr: 1.0e-05
48
+ smooth_dr: 1.0e-05
49
+ optimizer:
50
+ _target_: torch.optim.SGD
51
+ params: "[email protected]()"
52
+ momentum: 0.9
53
+ weight_decay: 4.0e-05
54
+ lr: 0.025
55
+ lr_scheduler:
56
+ _target_: torch.optim.lr_scheduler.StepLR
57
+ optimizer: "@optimizer"
58
+ step_size: 80
59
+ gamma: 0.5
60
+ image_key: image
61
+ label_key: label
62
+ val_interval: 10
63
+ train:
64
+ deterministic_transforms:
65
+ - _target_: LoadImaged
66
+ keys:
67
+ - "@image_key"
68
+ - "@label_key"
69
+ image_only: false
70
+ - _target_: EnsureChannelFirstd
71
+ keys:
72
+ - "@image_key"
73
+ - "@label_key"
74
+ - _target_: Orientationd
75
+ keys:
76
+ - "@image_key"
77
+ - "@label_key"
78
+ axcodes: RAS
79
+ - _target_: Spacingd
80
+ keys:
81
+ - "@image_key"
82
+ - "@label_key"
83
+ pixdim:
84
+ - 1
85
+ - 1
86
+ - 1
87
+ mode:
88
+ - bilinear
89
+ - nearest
90
+ align_corners:
91
+ - true
92
+ - true
93
+ - _target_: CastToTyped
94
+ keys: "@image_key"
95
+ dtype: "$torch.float32"
96
+ - _target_: ScaleIntensityRanged
97
+ keys: "@image_key"
98
+ a_min: -500
99
+ a_max: 500
100
+ b_min: 0
101
+ b_max: 1
102
+ clip: true
103
+ - _target_: CastToTyped
104
+ keys:
105
+ - "@image_key"
106
+ - "@label_key"
107
+ dtype:
108
+ - "$np.float16"
109
+ - "$np.uint8"
110
+ - _target_: CopyItemsd
111
+ keys: "@label_key"
112
+ times: 1
113
+ names:
114
+ - label4crop
115
+ - _target_: Lambdad
116
+ keys: label4crop
117
+ func: "$lambda x, s=@output_classes: np.concatenate(tuple([ndimage.binary_dilation((x==_k).astype(x.dtype),
118
+ iterations=48).astype(float) for _k in range(s)]), axis=0)"
119
+ overwrite: true
120
+ - _target_: EnsureTyped
121
+ keys:
122
+ - "@image_key"
123
+ - "@label_key"
124
+ - _target_: CastToTyped
125
+ keys: "@image_key"
126
+ dtype: "$torch.float32"
127
+ - _target_: SpatialPadd
128
+ keys:
129
+ - "@image_key"
130
+ - "@label_key"
131
+ - label4crop
132
+ spatial_size:
133
+ - 96
134
+ - 96
135
+ - 96
136
+ mode:
137
+ - reflect
138
+ - constant
139
+ - constant
140
+ random_transforms:
141
+ - _target_: RandCropByLabelClassesd
142
+ keys:
143
+ - "@image_key"
144
+ - "@label_key"
145
+ label_key: label4crop
146
+ num_classes: "@output_classes"
147
+ ratios: "$[1,] * @output_classes"
148
+ spatial_size:
149
+ - 96
150
+ - 96
151
+ - 96
152
+ num_samples: 1
153
+ - _target_: Lambdad
154
+ keys: label4crop
155
+ func: "$lambda x: 0"
156
+ - _target_: RandRotated
157
+ keys:
158
+ - "@image_key"
159
+ - "@label_key"
160
+ range_x: 0.3
161
+ range_y: 0.3
162
+ range_z: 0.3
163
+ mode:
164
+ - bilinear
165
+ - nearest
166
+ prob: 0.2
167
+ - _target_: RandZoomd
168
+ keys:
169
+ - "@image_key"
170
+ - "@label_key"
171
+ min_zoom: 0.8
172
+ max_zoom: 1.2
173
+ mode:
174
+ - trilinear
175
+ - nearest
176
+ align_corners:
177
+ - true
178
+ -
179
+ prob: 0.16
180
+ - _target_: RandGaussianSmoothd
181
+ keys: "@image_key"
182
+ sigma_x:
183
+ - 0.5
184
+ - 1.15
185
+ sigma_y:
186
+ - 0.5
187
+ - 1.15
188
+ sigma_z:
189
+ - 0.5
190
+ - 1.15
191
+ prob: 0.15
192
+ - _target_: RandScaleIntensityd
193
+ keys: "@image_key"
194
+ factors: 0.3
195
+ prob: 0.5
196
+ - _target_: RandShiftIntensityd
197
+ keys: "@image_key"
198
+ offsets: 0.1
199
+ prob: 0.5
200
+ - _target_: RandGaussianNoised
201
+ keys: "@image_key"
202
+ std: 0.01
203
+ prob: 0.15
204
+ - _target_: RandFlipd
205
+ keys:
206
+ - "@image_key"
207
+ - "@label_key"
208
+ spatial_axis: 0
209
+ prob: 0.5
210
+ - _target_: RandFlipd
211
+ keys:
212
+ - "@image_key"
213
+ - "@label_key"
214
+ spatial_axis: 1
215
+ prob: 0.5
216
+ - _target_: RandFlipd
217
+ keys:
218
+ - "@image_key"
219
+ - "@label_key"
220
+ spatial_axis: 2
221
+ prob: 0.5
222
+ - _target_: CastToTyped
223
+ keys:
224
+ - "@image_key"
225
+ - "@label_key"
226
+ dtype:
227
+ - "$torch.float32"
228
+ - "$torch.uint8"
229
+ - _target_: ToTensord
230
+ keys:
231
+ - "@image_key"
232
+ - "@label_key"
233
+ preprocessing:
234
+ _target_: Compose
235
+ transforms: "$@train#deterministic_transforms + @train#random_transforms"
236
+ dataset:
237
+ _target_: CacheDataset
238
+ data: "@train_datalist"
239
+ transform: "@train#preprocessing"
240
+ cache_rate: 0.125
241
+ num_workers: 4
242
+ dataloader:
243
+ _target_: DataLoader
244
+ dataset: "@train#dataset"
245
+ batch_size: 2
246
+ shuffle: true
247
+ num_workers: 4
248
+ inferer:
249
+ _target_: SimpleInferer
250
+ postprocessing:
251
+ _target_: Compose
252
+ transforms:
253
+ - _target_: Activationsd
254
+ keys: pred
255
+ softmax: true
256
+ - _target_: AsDiscreted
257
+ keys:
258
+ - pred
259
+ - label
260
+ argmax:
261
+ - true
262
+ - false
263
+ to_onehot: "@output_classes"
264
+ handlers:
265
+ - _target_: LrScheduleHandler
266
+ lr_scheduler: "@lr_scheduler"
267
+ print_lr: true
268
+ - _target_: ValidationHandler
269
+ validator: "@validate#evaluator"
270
+ epoch_level: true
271
+ interval: "@val_interval"
272
+ - _target_: StatsHandler
273
+ tag_name: train_loss
274
+ output_transform: "$monai.handlers.from_engine(['loss'], first=True)"
275
+ - _target_: TensorBoardStatsHandler
276
+ log_dir: "@output_dir"
277
+ tag_name: train_loss
278
+ output_transform: "$monai.handlers.from_engine(['loss'], first=True)"
279
+ key_metric:
280
+ train_accuracy:
281
+ _target_: ignite.metrics.Accuracy
282
+ output_transform: "$monai.handlers.from_engine(['pred', 'label'])"
283
+ trainer:
284
+ _target_: SupervisedTrainer
285
+ max_epochs: 400
286
+ device: "@device"
287
+ train_data_loader: "@train#dataloader"
288
+ network: "@network"
289
+ loss_function: "@loss"
290
+ optimizer: "@optimizer"
291
+ inferer: "@train#inferer"
292
+ postprocessing: "@train#postprocessing"
293
+ key_train_metric: "@train#key_metric"
294
+ train_handlers: "@train#handlers"
295
+ amp: true
296
+ validate:
297
+ preprocessing:
298
+ _target_: Compose
299
+ transforms: "%train#deterministic_transforms"
300
+ dataset:
301
+ _target_: CacheDataset
302
+ data: "@val_datalist"
303
+ transform: "@validate#preprocessing"
304
+ cache_rate: 0.125
305
+ dataloader:
306
+ _target_: DataLoader
307
+ dataset: "@validate#dataset"
308
+ batch_size: 1
309
+ shuffle: false
310
+ num_workers: 4
311
+ inferer:
312
+ _target_: SlidingWindowInferer
313
+ roi_size:
314
+ - 96
315
+ - 96
316
+ - 96
317
+ sw_batch_size: 6
318
+ overlap: 0.625
319
+ postprocessing: "%train#postprocessing"
320
+ handlers:
321
+ - _target_: StatsHandler
322
+ iteration_log: false
323
+ - _target_: TensorBoardStatsHandler
324
+ log_dir: "@output_dir"
325
+ iteration_log: false
326
+ - _target_: CheckpointSaver
327
+ save_dir: "@ckpt_dir"
328
+ save_dict:
329
+ model: "@network"
330
+ save_key_metric: true
331
+ key_metric_filename: model.pt
332
+ key_metric:
333
+ val_mean_dice:
334
+ _target_: MeanDice
335
+ include_background: false
336
+ output_transform: "$monai.handlers.from_engine(['pred', 'label'])"
337
+ additional_metrics:
338
+ val_accuracy:
339
+ _target_: ignite.metrics.Accuracy
340
+ output_transform: "$monai.handlers.from_engine(['pred', 'label'])"
341
+ evaluator:
342
+ _target_: SupervisedEvaluator
343
+ device: "@device"
344
+ val_data_loader: "@validate#dataloader"
345
+ network: "@network"
346
+ inferer: "@validate#inferer"
347
+ postprocessing: "@validate#postprocessing"
348
+ key_val_metric: "@validate#key_metric"
349
+ additional_metrics: "@validate#additional_metrics"
350
+ val_handlers: "@validate#handlers"
351
+ amp: true
352
+ initialize:
353
+ - "$monai.utils.set_determinism(seed=123)"
354
+ - "$setattr(torch.backends.cudnn, 'benchmark', True)"
355
+ run:
356
+ - "$@train#trainer.run()"
docs/README.md ADDED
@@ -0,0 +1,125 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Multi-organ segmentation in abdominal CT
2
+
3
+ ### **Authors**
4
+
5
+ Chen Shen<sup>1</sup>, Holger R. Roth<sup>2</sup>, Kazunari Misawa<sup>3</sup>, Kensaku Mori<sup>1</sup>
6
+
7
+ 1. Nagoya University, Japan
8
+
9
+ 2. NVIDIA Corporation, USA
10
+
11
+ 3. Aichi Cancer Center, Japan
12
+
13
+ ### **Tags**
14
+
15
+ Segmentation, Multi-organ, Abdominal
16
+
17
+ ## **Model Description**
18
+
19
+ This model uses the DiNTS model architecture searched on [Medical Segmentation Decathlon](http://medicaldecathlon.com/) Pancreas [1] and re-trained for multi-organ segmentation from abdominal CT images [2,3].
20
+
21
+ ## **Data**
22
+
23
+ This model was trained on an abdominal CT dataset in portal venous phase collected from Aichi Cancer Center in Japan. Since this is a private dataset, similar models can be trained using other public multi-organ datasets like [BTCV](https://www.synapse.org/#!Synapse:syn3193805/wiki/89480).
24
+
25
+ For this bundle, we split the 420 cases into training, validation and testing with 300, 60 and 60 cases, respectively.
26
+
27
+ ## **Output**
28
+ 8 channels
29
+
30
+ - 0: Background
31
+ - 1: Artery
32
+ - 2: Portal vein
33
+ - 3: Liver
34
+ - 4: Spleen
35
+ - 5: Stomach
36
+ - 6: Gallbladder
37
+ - 7: Pancreas
38
+
39
+ Here is an example of output.
40
+
41
+ ![alt用テキスト](output_example.png)
42
+
43
+ ## **Scores**
44
+
45
+ This model achieves the following Dice score on the validation data (our own split from the whole dataset):
46
+
47
+ Mean Dice = 88.6%
48
+
49
+ ## MONAI Bundle Commands
50
+ In addition to the Pythonic APIs, a few command line interfaces (CLI) are provided to interact with the bundle. The CLI supports flexible use cases, such as overriding configs at runtime and predefining arguments in a file.
51
+
52
+ For more details usage instructions, visit the [MONAI Bundle Configuration Page](https://docs.monai.io/en/latest/config_syntax.html).
53
+
54
+ #### Execute model searching:
55
+
56
+ ```
57
+ python -m scripts.search run --config_file configs/search.yaml
58
+ ```
59
+
60
+ #### Execute multi-GPU model searching (recommended):
61
+
62
+ ```
63
+ torchrun --nnodes=1 --nproc_per_node=8 -m scripts.search run --config_file configs/search.yaml
64
+ ```
65
+
66
+ #### Execute training:
67
+
68
+ ```
69
+ python -m monai.bundle run --config_file configs/train.yaml
70
+ ```
71
+
72
+ Please note that if the default dataset path is not modified with the actual path in the bundle config files, you can also override it by using `--dataset_dir`:
73
+
74
+ ```
75
+ python -m monai.bundle run --config_file configs/train.yaml
76
+ ```
77
+
78
+ #### Override the `train` config to execute multi-GPU training:
79
+
80
+ ```
81
+ torchrun --nnodes=1 --nproc_per_node=8 \
82
+ -m scripts.search run \
83
+ --config_file configs/search.yaml
84
+ ```
85
+
86
+ #### Override the `train` config to execute evaluation with the trained model:
87
+
88
+ ```
89
+ python -m monai.bundle run \
90
+ --config_file "['configs/train.yaml','configs/evaluate.yaml']"
91
+ ```
92
+
93
+ #### Execute inference:
94
+ ```
95
+ python -m monai.bundle run --config_file configs/inference.yaml
96
+ ```
97
+
98
+
99
+ #### Export checkpoint for TorchScript:
100
+
101
+ ```
102
+ python -m monai.bundle ckpt_export network_def --filepath models/model.ts --ckpt_file models/model.pt --meta_file configs/metadata.json --config_file configs/inference.yaml
103
+ ```
104
+
105
+ #### Execute inference with the TensorRT model:
106
+
107
+ ```
108
+ python -m monai.bundle run --config_file "['configs/inference.yaml', 'configs/inference_trt.yaml']"
109
+ ```
110
+
111
+
112
+ ## **References**
113
+
114
+ [1] He, Y., Yang, D., Roth, H., Zhao, C. and Xu, D., 2021. Dints: Differentiable neural network topology search for 3d medical image segmentation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 5841-5850).
115
+
116
+
117
+ [2] Roth, Holger R., et al. "A multi-scale pyramid of 3D fully convolutional networks for abdominal multi-organ segmentation." International conference on medical image computing and computer-assisted intervention. Springer, Cham, 2018.
118
+
119
+
120
+ [3] Shen, Chen, et al. "Effective hyperparameter optimization with proxy data for multi-organ segmentation." Medical Imaging 2022: Image Processing. Vol. 12032. SPIE, 2022.
121
+
122
+
123
+
124
+ ## **License**
125
+ The Licensee is not allowed to distribute or make the model to any third party, either for free or for a fee. Reverse engineering of the model is not allowed. This includes, but is not limited to, providing the model as part of a commercial offering, sharing the model on a public or private network, or making the model available for download on the Internet.
docs/output_example.png ADDED

Git LFS Details

  • SHA256: 0761759fbd84c9b545dd66f48ba592ca3cc2eff207581056fcfd56b83e97eb50
  • Pointer size: 131 Bytes
  • Size of remote file: 898 kB
models/model.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0ed27c74695d94a5f4b05f1f345b239ed3dcd630ee4dd4e7f593d7d7a740ca3b
3
+ size 616156645
models/model.ts ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3da668b2484e263ff01c148ed4210df4421f8bbd7683da34b7930d965011a9e5
3
+ size 616352855
models/search_code_18590.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:01e361e9843e2f4e5ff1599da0abac77013ea38cab8fdd6c9286bb6572c9a32d
3
+ size 4335