Datasets:

Modalities:
Image
Text
Formats:
parquet
ArXiv:
Libraries:
Datasets
Dask
License:
a-huli commited on
Commit
5046f16
1 Parent(s): 09f3d41

Upload 3 files

Browse files
Files changed (3) hide show
  1. labels.py +393 -0
  2. masks.py +106 -0
  3. pos_weights.json +1 -0
labels.py ADDED
@@ -0,0 +1,393 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """
2
+ All target labels used in UMIE.
3
+
4
+ Each label has a unique id, RadLex name, RadLex id, grade, and source names.
5
+ Source names are the names of the labels used in the original datasets. They are meant to be helpful when translating labels from new source datasets to RadLex labels.
6
+ Each of the target labels has a grade, which is an integer value that represents the intensity of the label ranging from 0 to 1.
7
+ E.g. 0.25 - mild, 0.5 - moderate, 0.75 - severe, 1 - critical. If the grade is not specified, the default value is 1.
8
+ Here in "grades", we store the information about how many grades the label has.
9
+ """
10
+
11
+ from dataclasses import dataclass, field
12
+ from typing import Optional
13
+
14
+
15
+ @dataclass
16
+ class Label: # Name of the label should match RadLex name
17
+ """This class represents a target UMIE label."""
18
+
19
+ id: int # Unique id of the label in UMIE
20
+ radlex_name: str # name of the label in RadLex
21
+ radlex_id: str # unique id of the label in RadLex
22
+ grades: Optional[int] = field( # number of grades the label has (possible intensity levels of the label)
23
+ default_factory=int
24
+ ) # Grades of the label, some datasets come with intesity scale for the labels
25
+ source_names: Optional[dict] = field(
26
+ default_factory=dict
27
+ ) # Names used by other datasets, key - dataset name, value - source label name in this dataset
28
+
29
+
30
+ # Use camel case for the labels names
31
+ # We do not keep vague labels like "other" or "unknown"
32
+ # One source label may correspond to multiple RadLex labels
33
+
34
+ NormalityDecriptor = Label(
35
+ id=0,
36
+ radlex_name="NormalityDecriptor",
37
+ radlex_id="RID29001",
38
+ source_names={
39
+ "kits23": ["normal"],
40
+ "coronahack": ["Normal"],
41
+ "alzheimers": ["NonDemented"],
42
+ "covid19_detection": ["normal"],
43
+ "chestX-ray14": ["No Finding"],
44
+ "brain_tumor_classification": ["no_tumor"],
45
+ "brain_with_intracranial_hemorrhage": ["normal"],
46
+ "brain_tumor_detection": ["N"],
47
+ "lits": ["NormalityDescriptor"],
48
+ },
49
+ )
50
+
51
+ Neoplasm = Label(
52
+ id=1,
53
+ radlex_name="Neoplasm",
54
+ radlex_id="RID3957",
55
+ source_names={
56
+ "kits23": [
57
+ "angiomyolipoma",
58
+ "chromophobe",
59
+ "clear_cell_papillary_rcc",
60
+ "clear_cell_rcc",
61
+ "mest",
62
+ "multilocular_cystic_rcc",
63
+ "oncocytoma",
64
+ "papillary_rcc",
65
+ "rcc_unclassified",
66
+ "spindle_cell_neoplasm",
67
+ "urothelial",
68
+ "wilms",
69
+ "other",
70
+ ],
71
+ "pad_chest": ["adenocarcinoma"],
72
+ "brain_tumor_detection": ["Y"],
73
+ "lits": ["Neoplasm"],
74
+ },
75
+ )
76
+
77
+ RenalAdenocarcinoma = Label(
78
+ id=2,
79
+ radlex_name="RenalAdenocarcinoma",
80
+ radlex_id="RID4234",
81
+ source_names={
82
+ "kits23": [
83
+ "clear_cell_rcc",
84
+ "chromophobe_rcc",
85
+ "papillary_rcc",
86
+ "multilocular_cystic_rcc",
87
+ "rcc_unclassified",
88
+ "clear_cell_papillary",
89
+ ],
90
+ },
91
+ )
92
+
93
+ ClearCellAdenocarcinoma = Label(
94
+ id=3,
95
+ radlex_name="ClearCellAdenocarcinoma",
96
+ radlex_id="RID4235",
97
+ source_names={"kits23": ["clear_cell_rcc", "clear_cell_papillary"]},
98
+ )
99
+
100
+ ChromophobeAdenocarcinoma = Label(
101
+ id=4, radlex_name="ChromophobeAdenocarcinoma", radlex_id="RID4236", source_names={"kits23": ["chromophobe_rcc"]}
102
+ )
103
+
104
+ TransitionalCellCarcinoma = Label(
105
+ id=5,
106
+ radlex_name="TransitionalCellCarcinoma",
107
+ radlex_id="",
108
+ source_names={"kits23": ["transitional_cell_carcinoma"]},
109
+ )
110
+
111
+ PapillaryRenalAdenocarcinoma = Label(
112
+ id=6,
113
+ radlex_name="PapillaryRenalAdenocarcinoma",
114
+ radlex_id="RID4233",
115
+ source_names={"kits23": ["papillary_rcc", "clear_cell_papillary"]},
116
+ )
117
+
118
+ MultilocularCysticRenalTumor = Label(
119
+ id=7,
120
+ radlex_name="MultilocularCysticRenalTumor",
121
+ radlex_id="RID4538",
122
+ source_names={"kits23": ["multilocular_cystic_rcc"]},
123
+ )
124
+
125
+ WilmsTumor = Label(id=8, radlex_name="WilmsTumor", radlex_id="RID4553", source_names={"kits23": ["wilms"]})
126
+
127
+ Angiomyolipoma = Label(
128
+ id=9, radlex_name="Angiomyolipoma", radlex_id="RID4343", source_names={"kits23": ["angiomyolipoma"]}
129
+ )
130
+
131
+ Oncocytoma = Label(id=10, radlex_name="Oncocytoma", radlex_id="RID4515", source_names={"kits23": ["oncocytoma"]})
132
+
133
+ RenalCyst = Label(id=11, radlex_name="RenalCyst", radlex_id="RID35811", source_names={"kits23": ["cyst"]})
134
+
135
+ ViralInfection = Label(
136
+ id=12,
137
+ radlex_name="ViralInfection",
138
+ radlex_id="RID4687",
139
+ grades=5,
140
+ source_names={
141
+ "coronahack": ["PneumoniaVirus"],
142
+ "MosMedData": ["CT-1", "CT-2", "CT-3", "CT-4"],
143
+ "covid19_detection": ["pneumonia_viral"],
144
+ },
145
+ )
146
+
147
+ Pneumonia = Label(
148
+ id=13,
149
+ radlex_name="Pneumonia",
150
+ radlex_id="RID5350",
151
+ source_names={
152
+ "coronahack": ["PneumoniaVirus", "PneumoniaBacteria"],
153
+ "ChestX-ray14": ["Pneumonia"],
154
+ "PadChest": ["Pneumonia", "atypical pneumonia"],
155
+ "covid19_detection": ["pneumonia_bacterial", "pneumonia_viral"],
156
+ },
157
+ )
158
+
159
+ PneumoniaViral = Label(
160
+ id=14,
161
+ radlex_name="PneumoniaViral",
162
+ radlex_id="RID34769",
163
+ source_names={"coronahack": ["PneumoniaVirus"], "covid19_detection": ["pneumonia_viral"]},
164
+ )
165
+
166
+ Atelectasis = Label(
167
+ id=15,
168
+ radlex_name="Atelectasis",
169
+ radlex_id="RID28493",
170
+ source_names={"Chest_X-ray_Abnormalities_Detection": ["Atelectasis"], "ChestX-ray14": ["Atelectasis"]},
171
+ )
172
+
173
+ Calcification = Label(
174
+ id=16,
175
+ radlex_name="Calcification",
176
+ radlex_id="RID5196",
177
+ source_names={"Chest_X-ray_Abnormalities_Detection": ["Calcification"]},
178
+ )
179
+
180
+ BoxlikeHeart = Label(
181
+ id=17,
182
+ radlex_name="BoxlikeHeart",
183
+ radlex_id="RID35057",
184
+ source_names={"Chest_X-ray_Abnormalities_Detection": ["Cardiomegaly"], "ChestX-ray14": ["Cardiomegaly"]},
185
+ )
186
+
187
+ Consolidation = Label(
188
+ id=18,
189
+ radlex_name="Consolidation",
190
+ radlex_id="RID43255",
191
+ source_names={
192
+ "Chest_X-ray_Abnormalities_Detection": ["Consolidation", "Infiltration"],
193
+ "ChestX-ray14": ["Consolidation", "Infiltration"],
194
+ },
195
+ )
196
+
197
+ InterstitialLungDisease = Label(
198
+ id=19,
199
+ radlex_name="InterstitialLungDisease",
200
+ radlex_id="RID28799",
201
+ source_names={"Chest_X-ray_Abnormalities_Detection": ["ILD"]},
202
+ )
203
+
204
+ Opacity = Label(
205
+ id=20,
206
+ radlex_name="Opacity",
207
+ radlex_id="RID28530",
208
+ source_names={"Chest_X-ray_Abnormalities_Detection": ["Opacity"]},
209
+ )
210
+
211
+ Lesion = Label(
212
+ id=21,
213
+ radlex_name="Lesion",
214
+ radlex_id="RID38780",
215
+ source_names={
216
+ "Chest_X-ray_Abnormalities_Detection": ["Nodule/Mass", "Other Lesion"],
217
+ "ChestX-ray14": ["Mass", "Nodule"],
218
+ },
219
+ )
220
+
221
+ PleuralEffusion = Label(
222
+ id=22,
223
+ radlex_name="PleuralEffusion",
224
+ radlex_id="RID34539",
225
+ source_names={"Chest_X-ray_Abnormalities_Detection": ["Effusion"], "ChestX-ray14": ["Effusion"]},
226
+ )
227
+
228
+ Thickening = Label(
229
+ id=23,
230
+ radlex_name="Thickening",
231
+ radlex_id="RID5352",
232
+ source_names={
233
+ "Chest_X-ray_Abnormalities_Detection": ["Pleural_Thickening"],
234
+ "ChestX-ray14": ["Pleural_Thickening"],
235
+ },
236
+ )
237
+
238
+ Pneumothorax = Label(
239
+ id=24,
240
+ radlex_name="Pneumothorax",
241
+ radlex_id="RID5352",
242
+ source_names={"Chest_X-ray_Abnormalities_Detection": ["Pneumothorax"], "ChestX-ray14": ["Pneumothorax"]},
243
+ )
244
+
245
+ Fibrosis = Label(
246
+ id=25,
247
+ radlex_name="Fibrosis",
248
+ radlex_id="RID3820",
249
+ source_names={
250
+ "Chest_X-ray_Abnormalities_Detection": ["Pulmonary Fibrosis"],
251
+ "ChestX-ray14": ["Fibrosis"],
252
+ "PadChest": ["Pulmonary Fibrosis"],
253
+ },
254
+ )
255
+
256
+ Mass = Label(id=26, radlex_name="Mass", radlex_id="RID3874", source_names={"ChestX-ray14": ["Mass"]})
257
+
258
+ PulmonaryEdema = Label(
259
+ id=27,
260
+ radlex_name="PulmonaryEdema",
261
+ radlex_id="RID4866",
262
+ source_names={"ChestX-ray14": ["Edema"], "PadChest": ["Pulmonary Edema"]},
263
+ )
264
+
265
+ Emphysema = Label(
266
+ id=28,
267
+ radlex_name="Emphysema",
268
+ radlex_id="RID4799",
269
+ source_names={"ChestX-ray14": ["Emphysema"], "PadChest": ["Emphysema"]},
270
+ )
271
+
272
+ Hernia = Label(id=29, radlex_name="Hernia", radlex_id="RID4895", source_names={"ChestX-ray14": ["Hernia"]})
273
+
274
+ ChronicObstructivePulmonaryDisease = Label(
275
+ id=30,
276
+ radlex_name="ChronicObstructivePulmonaryDisease",
277
+ radlex_id="RID5317",
278
+ source_names={"PadChest": ["COPD signs"]},
279
+ )
280
+
281
+ Tubeculosis = Label(
282
+ id=31,
283
+ radlex_name="Tuberculosis",
284
+ radlex_id="RID29116",
285
+ source_names={"PadChest": ["Tuberculosis", "Tuberculosis seqelae"]},
286
+ )
287
+
288
+ Metastasis = Label(
289
+ id=32,
290
+ radlex_name="Metastasis",
291
+ radlex_id="RID5231",
292
+ source_names={"PadChest": ["Lung metastasis", "Bone metastasis"]},
293
+ )
294
+
295
+ Pneumonitis = Label(
296
+ id=33, radlex_name="Pneumonitis", radlex_id="RID3541", source_names={"PadChest": ["post radiotherapy changes"]}
297
+ )
298
+
299
+ PulmonaryHypertension = Label(
300
+ id=34,
301
+ radlex_name="PulmonaryHypertension",
302
+ radlex_id="RID3299",
303
+ source_names={"PadChest": ["Pulmonary artery hypertension"]},
304
+ )
305
+
306
+ AdultRespiratoryDistressSyndrome = Label(
307
+ id=35,
308
+ radlex_name="AdultRespiratoryDistressSyndrome",
309
+ radlex_id="RID5319",
310
+ source_names={"PadChest": ["Respiratory distress syndrome"]},
311
+ )
312
+
313
+ Asbestosis = Label(
314
+ id=36, radlex_name="Asbestosis", radlex_id="RID5346", source_names={"PadChest": {"Asbestosis signs"}}
315
+ )
316
+
317
+ Carcinomatosis = Label(
318
+ id=37, radlex_name="Carcinomatosis", radlex_id="RID5231", source_names={"PadChest": ["lymphangitis carcinomatosa"]}
319
+ )
320
+
321
+ Adenocarcinoma = Label(
322
+ id=38,
323
+ radlex_name="Adenocarcinoma",
324
+ radlex_id="RID4226",
325
+ source_names={
326
+ "kits23": [
327
+ "clear_cell_rcc",
328
+ "chromophobe_rcc",
329
+ "papillary_rcc",
330
+ "multilocular_cystic_rcc",
331
+ "rcc_unclassified",
332
+ "clear_cell_papillary",
333
+ ],
334
+ "PadChest": ["adenocarcinoma"],
335
+ },
336
+ )
337
+
338
+ Glioma = Label(
339
+ id=39, radlex_name="Glioma", radlex_id="RID4026", source_names={"brain_tumor_classification": ["glioma_tumor"]}
340
+ )
341
+
342
+ Meningioma = Label(
343
+ id=40,
344
+ radlex_name="Meningioma",
345
+ radlex_id="RID4088",
346
+ source_names={"brain_tumor_classification": ["meningioma_tumor"]},
347
+ )
348
+
349
+ Pituitary = Label(
350
+ id=41,
351
+ radlex_name="Pituitary",
352
+ radlex_id="RID28679",
353
+ source_names={"brain_tumor_classification": ["pituitary_tumor"]},
354
+ )
355
+
356
+ Osteoarthritis = Label(
357
+ id=42,
358
+ radlex_name="Osteoarthritis",
359
+ radlex_id="RID3555",
360
+ grades=4,
361
+ source_names={
362
+ "knee_osteoarthritis": [
363
+ "DoubtfulOsteoarthritis",
364
+ "MinimalOsteoarthritis",
365
+ "ModerateOsteoarthritis",
366
+ "SevereOsteoarthritis",
367
+ ]
368
+ },
369
+ )
370
+
371
+ Hemorrhage = Label(
372
+ id=43,
373
+ radlex_name="Hemorrhage",
374
+ radlex_id="RID4700",
375
+ source_names={"Brain_with_hemorrhage": ["brain_hemorrhage"]},
376
+ )
377
+
378
+ Dementia = Label(
379
+ id=44,
380
+ radlex_name="Dementia",
381
+ radlex_id="RID5136",
382
+ grades=3,
383
+ source_names={"alzheimers": ["VeryMildDemented", "MildDemented", "ModerateDemented"]},
384
+ )
385
+
386
+ HeartFailure = Label(
387
+ id=45,
388
+ radlex_name="HeartFailure",
389
+ radlex_id="RID34795",
390
+ source_names={"PadChest": ["heart insufficiency"]},
391
+ )
392
+
393
+ all_labels = [obj for name, obj in globals().items() if isinstance(obj, Label)]
masks.py ADDED
@@ -0,0 +1,106 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ """
2
+ This file contains the masks used in the datasets. The masks are defined as mask dataclasses.
3
+
4
+ Each mask has the following attributes:
5
+ - id: int - the id of the mask
6
+ - radlex_name: str - the name of the mask in RadLex
7
+ - color: int - the color of the mask
8
+ - radlex_id: str - the id of the mask in RadLex
9
+ - source_names: dict - the names of the mask in other datasets, meant to be helpful when translating new source dataset masks to UMIE.
10
+ """
11
+ from dataclasses import dataclass, field
12
+ from typing import Optional
13
+
14
+
15
+ @dataclass
16
+ class Mask:
17
+ """This class represents a mask."""
18
+
19
+ id: int # The unique id of the mask in UMIE (also the value of the mask in the mask image)
20
+ radlex_name: str # The name of the mask in RadLex
21
+ color: int # The color of the mask in the mask image
22
+ radlex_id: str # The id of the mask in RadLex
23
+ source_names: Optional[dict] = field(
24
+ default_factory=dict
25
+ ) # How the mask is named in other datasets, key - dataset name, value - source mask name in this dataset
26
+
27
+
28
+ Background = Mask(
29
+ id=0, radlex_name="Background", color=0, radlex_id=""
30
+ ) # Actually there is no radlex for backgroud ofc but "Normality descriptor" didnt fit either
31
+
32
+ Kidney = Mask(
33
+ id=1, radlex_name="Kidney", color=1, radlex_id="RID205", source_names={"kits23": ["kidney"], "ct_org": ["Kidney"]}
34
+ )
35
+
36
+ Neoplasm = Mask(
37
+ id=2,
38
+ radlex_name="Neoplasm",
39
+ color=2,
40
+ radlex_id="RID3957",
41
+ source_names={"kits23": ["kidney tumor"], "brain_tumor_progression": ["Brain tumor"], "lits": ["liver_tumor"]},
42
+ )
43
+ RenalCyst = Mask(id=3, radlex_name="RenalCyst", color=3, radlex_id="RID35811", source_names={"kits23": ["cyst"]})
44
+ ViralInfection = Mask(
45
+ id=4, radlex_name="ViralInfection", color=4, radlex_id="RID4687", source_names={"mos_med_data": ["CT-1"]}
46
+ )
47
+
48
+ Lung = Mask(
49
+ id=5,
50
+ radlex_name="Lung",
51
+ color=5,
52
+ radlex_id="RID1301",
53
+ source_names={
54
+ "ct_org": ["lungs"],
55
+ "chest_xray_masks_and_label": ["Lungs"],
56
+ "finding_and_measuring_lungs": ["lungs"],
57
+ },
58
+ )
59
+
60
+ BoneOrgan = Mask(id=6, radlex_name="BoneOrgan", color=6, radlex_id="RID13197", source_names={"ct_org": ["Bones"]})
61
+
62
+ Liver = Mask(id=7, radlex_name="Liver", color=7, radlex_id="RID58", source_names={"ct_org": ["liver"]})
63
+
64
+ UrinaryBladder = Mask(
65
+ id=8, radlex_name="UrinaryBladder", color=8, radlex_id="RID237", source_names={"ct_org": ["bladder"]}
66
+ )
67
+
68
+ Brain = Mask(id=9, radlex_name="Brain", color=9, radlex_id="RID6434", source_names={"ct_org": ["Brain"]})
69
+
70
+ Nodule = Mask(
71
+ id=10,
72
+ radlex_name="Nodule",
73
+ color=10,
74
+ radlex_id="RID3875",
75
+ source_names={"lidc_idri": ["Nodule>=3mm", "Nodule<3mm"]},
76
+ )
77
+
78
+ Lesion = Mask(
79
+ id=11, radlex_name="Lesion", color=11, radlex_id="RID38780", source_names={"lidc_idri": ["Non-nodule>=3mm"]}
80
+ ) # That is not a nodule
81
+
82
+ CalciumScore = Mask(
83
+ id=12,
84
+ radlex_name="CalciumScore",
85
+ color=12,
86
+ radlex_id="RID28808",
87
+ source_names={"coca": ["coronary_artery_calcium"]},
88
+ )
89
+
90
+ Metastasis = Mask(
91
+ id=13,
92
+ radlex_name="Metastasis",
93
+ color=13,
94
+ radlex_id="RID5231",
95
+ source_names={"brain_met_share": ["brain_metastasis"]},
96
+ )
97
+
98
+ Hemorrhage = Mask(
99
+ id=14,
100
+ radlex_name="Hemorrhage",
101
+ color=14,
102
+ radlex_id="RID4700",
103
+ source_names={"brain_with_intracranial_hemorrhage": ["brain_hemorrhage"]},
104
+ )
105
+
106
+ all_masks = [obj for name, obj in globals().items() if isinstance(obj, Mask)]
pos_weights.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"label_counts": {"NormalityDecriptor": 1.1057690645367633, "Neoplasm": 0.09583273355405211, "RenalAdenocarcinoma": 0.054858642325462566, "ClearCellAdenocarcinoma": 0.04114881613813146, "ChromophobeAdenocarcinoma": 0.00594213132151514, "TransitionalCellCarcinoma": 0, "PapillaryRenalAdenocarcinoma": 0.007627272877682368, "MultilocularCysticRenalTumor": 9.03171775060964e-05, "WilmsTumor": 0, "Angiomyolipoma": 0.0003120739779577222, "Oncocytoma": 0.004200520221444501, "RenalCyst": 0, "ViralInfection": 0.01151410716880852, "Pneumonia": 0.03661655383122769, "PneumoniaViral": 0.01151410716880852, "Atelectasis": 0.04981275506466307, "Calcification": 0, "BoxlikeHeart": 0.011526707414297104, "Consolidation": 0.11212662122740781, "InterstitialLungDisease": 0, "Opacity": 0, "Lesion": 0.052325104213913905, "PleuralEffusion": 0.05782683648080038, "Thickening": 0.014091073710677162, "Pneumothorax": 0.022248705445939255, "Fibrosis": 0.0069691884161010575, "Mass": 0.024311891887346213, "PulmonaryEdema": 0.009543938169536479, "Emphysema": 0.010435850214855739, "Hernia": 0.0009326940065165316, "ChronicObstructivePulmonaryDisease": 0, "Tuberculosis": 0, "Metastasis": 0, "Pneumonitis": 0, "PulmonaryHypertension": 0, "AdultRespiratoryDistressSyndrome": 0, "Asbestosis": 0, "Carcinomatosis": 0, "Adenocarcinoma": 0.054858642325462566, "Glioma": 0.0038156929644555426, "Meningioma": 0.003861194786356837, "Pituitary": 0.003712295071835588, "Osteoarthritis": 0.024945409565001537, "Hemorrhage": 0.002617585565414945, "Dementia": 0.013310705134604505, "HeartFailure": 0}, "mask_counts": {"Background": 0, "Kidney": 0.1532667717638816, "Neoplasm": 0.08602303933806484, "RenalCyst": 0.019574688926928693, "ViralInfection": 0, "Lung": 0.001097225703847687, "BoneOrgan": 0, "Liver": 0.08524893973413165, "UrinaryBladder": 0, "Brain": 0, "Nodule": 0, "Lesion": 0, "CalciumScore": 0.015244842675557408, "Metastasis": 0, "Hemorrhage": 0.002617585565414945}, "label_counts_list": [1.1057690645367633, 0.09583273355405211, 0.054858642325462566, 0.04114881613813146, 0.00594213132151514, 0, 0.007627272877682368, 9.03171775060964e-05, 0, 0.0003120739779577222, 0.004200520221444501, 0, 0.01151410716880852, 0.03661655383122769, 0.01151410716880852, 0.04981275506466307, 0, 0.011526707414297104, 0.11212662122740781, 0, 0, 0.052325104213913905, 0.05782683648080038, 0.014091073710677162, 0.022248705445939255, 0.0069691884161010575, 0.024311891887346213, 0.009543938169536479, 0.010435850214855739, 0.0009326940065165316, 0, 0, 0, 0, 0, 0, 0, 0, 0.054858642325462566, 0.0038156929644555426, 0.003861194786356837, 0.003712295071835588, 0.024945409565001537, 0.002617585565414945, 0.013310705134604505, 0], "mask_counts_list": [0, 0.1532667717638816, 0.08602303933806484, 0.019574688926928693, 0, 0.001097225703847687, 0, 0.08524893973413165, 0, 0, 0, 0, 0.015244842675557408, 0, 0.002617585565414945]}