Chris Oswald
commited on
Commit
·
5fa2de2
1
Parent(s):
f7a1cfb
generated examples
Browse files
SPIDER.py
CHANGED
|
@@ -14,17 +14,28 @@
|
|
| 14 |
# TODO: Address all TODOs and remove all explanatory comments
|
| 15 |
"""TODO: Add a description here."""
|
| 16 |
|
| 17 |
-
|
| 18 |
import csv
|
| 19 |
import json
|
| 20 |
import os
|
| 21 |
-
from typing import Dict, List, Optional, Set
|
| 22 |
|
| 23 |
import numpy as np
|
| 24 |
|
| 25 |
import datasets
|
| 26 |
import SimpleITK as sitk
|
| 27 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 28 |
# TODO: Add BibTeX citation
|
| 29 |
# Find for instance the citation on arxiv or on the dataset repo/website
|
| 30 |
_CITATION = """\
|
|
@@ -156,11 +167,36 @@ class SPIDER(datasets.GeneratorBasedBuilder):
|
|
| 156 |
]
|
| 157 |
|
| 158 |
# method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
| 159 |
-
def _generate_examples(
|
| 160 |
-
|
| 161 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 162 |
|
| 163 |
-
|
|
|
|
|
|
|
|
|
|
| 164 |
#TODO: make hardcoded values dynamic
|
| 165 |
np.random.seed(9999)
|
| 166 |
N_PATIENTS = 257
|
|
@@ -188,6 +224,21 @@ class SPIDER(datasets.GeneratorBasedBuilder):
|
|
| 188 |
overview_data = import_csv_data(paths_dict['overview'])
|
| 189 |
grades_data = import_csv_data(paths_dict['gradings'])
|
| 190 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 191 |
# Import image and mask data
|
| 192 |
image_files = [
|
| 193 |
file for file in os.listdir(os.path.join(paths_dict['images'], 'images'))
|
|
@@ -221,8 +272,10 @@ class SPIDER(datasets.GeneratorBasedBuilder):
|
|
| 221 |
subset_ids = test_ids
|
| 222 |
else:
|
| 223 |
subset_ids = None
|
| 224 |
-
raise ValueError(
|
| 225 |
-
|
|
|
|
|
|
|
| 226 |
|
| 227 |
image_files = [
|
| 228 |
file for file in image_files
|
|
@@ -258,40 +311,25 @@ class SPIDER(datasets.GeneratorBasedBuilder):
|
|
| 258 |
image_array = sitk.GetArrayFromImage(image)
|
| 259 |
|
| 260 |
# Extract overview data corresponding to image
|
| 261 |
-
|
| 262 |
-
# Extract patient radiological gradings corresponding to image
|
| 263 |
-
|
| 264 |
-
|
| 265 |
-
|
| 266 |
-
|
| 267 |
-
|
| 268 |
-
|
| 269 |
|
| 270 |
-
|
| 271 |
-
|
| 272 |
-
|
| 273 |
-
|
| 274 |
-
|
| 275 |
-
|
| 276 |
-
for line in reader:
|
| 277 |
-
results.append(line)
|
| 278 |
-
return results
|
| 279 |
-
|
| 280 |
-
|
| 281 |
|
| 282 |
-
|
| 283 |
-
|
| 284 |
-
|
| 285 |
-
|
| 286 |
-
|
| 287 |
-
|
| 288 |
-
|
| 289 |
-
|
| 290 |
-
|
| 291 |
-
|
| 292 |
-
|
| 293 |
-
|
| 294 |
-
|
| 295 |
-
"option2": data["option2"],
|
| 296 |
-
"second_domain_answer": "" if split == "test" else data["second_domain_answer"],
|
| 297 |
-
}
|
|
|
|
| 14 |
# TODO: Address all TODOs and remove all explanatory comments
|
| 15 |
"""TODO: Add a description here."""
|
| 16 |
|
| 17 |
+
# Import packages
|
| 18 |
import csv
|
| 19 |
import json
|
| 20 |
import os
|
| 21 |
+
from typing import Dict, List, Optional, Set, Tuple
|
| 22 |
|
| 23 |
import numpy as np
|
| 24 |
|
| 25 |
import datasets
|
| 26 |
import SimpleITK as sitk
|
| 27 |
|
| 28 |
+
# Define functions
|
| 29 |
+
def import_csv_data(filepath: str) -> List[Dict[str, str]]:
|
| 30 |
+
"""Import all rows of CSV file."""
|
| 31 |
+
results = []
|
| 32 |
+
with open(filepath, encoding='utf-8') as f:
|
| 33 |
+
reader = csv.DictReader(f)
|
| 34 |
+
for line in reader:
|
| 35 |
+
results.append(line)
|
| 36 |
+
return results
|
| 37 |
+
|
| 38 |
+
|
| 39 |
# TODO: Add BibTeX citation
|
| 40 |
# Find for instance the citation on arxiv or on the dataset repo/website
|
| 41 |
_CITATION = """\
|
|
|
|
| 167 |
]
|
| 168 |
|
| 169 |
# method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
|
| 170 |
+
def _generate_examples(
|
| 171 |
+
self,
|
| 172 |
+
paths_dict: Dict[str, str],
|
| 173 |
+
split: str = 'train', # ['train', 'validate', 'test']
|
| 174 |
+
validate_share: float = 0.3,
|
| 175 |
+
test_share: float = 0.2,
|
| 176 |
+
raw_image: bool = True,
|
| 177 |
+
numeric_array: bool = True,
|
| 178 |
+
metadata: bool = True,
|
| 179 |
+
rad_gradings: bool = True,
|
| 180 |
+
) -> Tuple[str, Dict]:
|
| 181 |
+
"""
|
| 182 |
+
This method handles input defined in _split_generators to yield
|
| 183 |
+
(key, example) tuples from the dataset. The `key` is for legacy reasons
|
| 184 |
+
(tfds) and is not important in itself, but must be unique for each example.
|
| 185 |
+
|
| 186 |
+
Args
|
| 187 |
+
paths_dict
|
| 188 |
+
split:
|
| 189 |
+
validate_share
|
| 190 |
+
test_share
|
| 191 |
+
raw_image
|
| 192 |
+
numeric_array
|
| 193 |
+
metadata
|
| 194 |
+
rad_gradings
|
| 195 |
|
| 196 |
+
Yields
|
| 197 |
+
|
| 198 |
+
"""
|
| 199 |
+
# Configure params
|
| 200 |
#TODO: make hardcoded values dynamic
|
| 201 |
np.random.seed(9999)
|
| 202 |
N_PATIENTS = 257
|
|
|
|
| 224 |
overview_data = import_csv_data(paths_dict['overview'])
|
| 225 |
grades_data = import_csv_data(paths_dict['gradings'])
|
| 226 |
|
| 227 |
+
# Convert overview data list of dicts to dict of dicts
|
| 228 |
+
overview_dict = {}
|
| 229 |
+
for item in overview_data:
|
| 230 |
+
key = item['new_file_name']
|
| 231 |
+
overview_dict[key] = item
|
| 232 |
+
|
| 233 |
+
# Merge patient records for radiological gradings data
|
| 234 |
+
grades_dict = {}
|
| 235 |
+
for patient_id in patient_ids:
|
| 236 |
+
patient_grades = [
|
| 237 |
+
x for x in grades_data if x['Patient'] == str(patient_id)
|
| 238 |
+
]
|
| 239 |
+
if patient_grades:
|
| 240 |
+
grades_dict[str(patient_id)] = patient_grades
|
| 241 |
+
|
| 242 |
# Import image and mask data
|
| 243 |
image_files = [
|
| 244 |
file for file in os.listdir(os.path.join(paths_dict['images'], 'images'))
|
|
|
|
| 272 |
subset_ids = test_ids
|
| 273 |
else:
|
| 274 |
subset_ids = None
|
| 275 |
+
raise ValueError( #TODO: move all parameter checking to beginning
|
| 276 |
+
f'Split argument "{split}" is not recognized. \
|
| 277 |
+
Please enter one of ["train", "validate", "test"]'
|
| 278 |
+
)
|
| 279 |
|
| 280 |
image_files = [
|
| 281 |
file for file in image_files
|
|
|
|
| 311 |
image_array = sitk.GetArrayFromImage(image)
|
| 312 |
|
| 313 |
# Extract overview data corresponding to image
|
| 314 |
+
image_overview = overview_dict[scan_id]
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 315 |
|
| 316 |
+
# Extract patient radiological gradings corresponding to image
|
| 317 |
+
patient_grades_dict = {}
|
| 318 |
+
for item in grades_dict[patient_id]:
|
| 319 |
+
key = f'IVD{item["IVD label"]}'
|
| 320 |
+
value = {k:v for k,v in item.items() if k not in ['Patient', 'IVD label']}
|
| 321 |
+
patient_grades_dict[key] = value
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 322 |
|
| 323 |
+
# Prepare example return dict
|
| 324 |
+
return_dict = {'patient_id':patient_id, 'scan_type':scan_type}
|
| 325 |
+
if raw_image:
|
| 326 |
+
return_dict['raw_image'] = image
|
| 327 |
+
if numeric_array:
|
| 328 |
+
return_dict['numeric_array'] = image_array
|
| 329 |
+
if metadata:
|
| 330 |
+
return_dict['metadata'] = image_overview
|
| 331 |
+
if rad_gradings:
|
| 332 |
+
return_dict['rad_gradings'] = patient_grades_dict
|
| 333 |
+
|
| 334 |
+
# Yield example
|
| 335 |
+
yield (scan_id, return_dict)
|
|
|
|
|
|
|
|
|