|
a |
|
b/Serialized/Prepare.ipynb |
|
|
1 |
{ |
|
|
2 |
"cells": [ |
|
|
3 |
{ |
|
|
4 |
"cell_type": "code", |
|
|
5 |
"execution_count": 1, |
|
|
6 |
"metadata": {}, |
|
|
7 |
"outputs": [ |
|
|
8 |
{ |
|
|
9 |
"name": "stderr", |
|
|
10 |
"output_type": "stream", |
|
|
11 |
"text": [ |
|
|
12 |
"/home/reina/anaconda3/envs/RSNA/lib/python3.6/importlib/_bootstrap.py:219: ImportWarning: can't resolve package from __spec__ or __package__, falling back on __name__ and __path__\n", |
|
|
13 |
" return f(*args, **kwds)\n", |
|
|
14 |
"/home/reina/anaconda3/envs/RSNA/lib/python3.6/importlib/_bootstrap.py:219: ImportWarning: can't resolve package from __spec__ or __package__, falling back on __name__ and __path__\n", |
|
|
15 |
" return f(*args, **kwds)\n" |
|
|
16 |
] |
|
|
17 |
} |
|
|
18 |
], |
|
|
19 |
"source": [ |
|
|
20 |
"from __future__ import absolute_import\n", |
|
|
21 |
"from __future__ import division\n", |
|
|
22 |
"from __future__ import print_function\n", |
|
|
23 |
"\n", |
|
|
24 |
"import glob, pylab, pandas as pd\n", |
|
|
25 |
"import pydicom, numpy as np\n", |
|
|
26 |
"from os import listdir\n", |
|
|
27 |
"from os.path import isfile, join\n", |
|
|
28 |
"import matplotlib.pylab as plt\n", |
|
|
29 |
"import os\n", |
|
|
30 |
"import seaborn as sns\n", |
|
|
31 |
"import warnings\n", |
|
|
32 |
"warnings.filterwarnings(action='once')\n", |
|
|
33 |
"import pickle\n", |
|
|
34 |
"from tqdm import tqdm, tqdm_notebook\n", |
|
|
35 |
"%load_ext autoreload\n", |
|
|
36 |
"%autoreload 2\n", |
|
|
37 |
"%matplotlib notebook\n", |
|
|
38 |
"from helper import *\n", |
|
|
39 |
"import time\n", |
|
|
40 |
"from IPython.core.interactiveshell import InteractiveShell\n", |
|
|
41 |
"InteractiveShell.ast_node_interactivity = \"all\"\n", |
|
|
42 |
"from defenitions import *" |
|
|
43 |
] |
|
|
44 |
}, |
|
|
45 |
{ |
|
|
46 |
"cell_type": "code", |
|
|
47 |
"execution_count": 65, |
|
|
48 |
"metadata": {}, |
|
|
49 |
"outputs": [], |
|
|
50 |
"source": [ |
|
|
51 |
"train_base_df = pd.read_csv(train_images_dir)" |
|
|
52 |
] |
|
|
53 |
}, |
|
|
54 |
{ |
|
|
55 |
"cell_type": "code", |
|
|
56 |
"execution_count": 67, |
|
|
57 |
"metadata": {}, |
|
|
58 |
"outputs": [ |
|
|
59 |
{ |
|
|
60 |
"data": { |
|
|
61 |
"text/plain": [ |
|
|
62 |
"4045572" |
|
|
63 |
] |
|
|
64 |
}, |
|
|
65 |
"execution_count": 67, |
|
|
66 |
"metadata": {}, |
|
|
67 |
"output_type": "execute_result" |
|
|
68 |
} |
|
|
69 |
], |
|
|
70 |
"source": [ |
|
|
71 |
"train_base_df.shape[0]" |
|
|
72 |
] |
|
|
73 |
}, |
|
|
74 |
{ |
|
|
75 |
"cell_type": "code", |
|
|
76 |
"execution_count": 4, |
|
|
77 |
"metadata": {}, |
|
|
78 |
"outputs": [ |
|
|
79 |
{ |
|
|
80 |
"data": { |
|
|
81 |
"text/html": [ |
|
|
82 |
"<div>\n", |
|
|
83 |
"<style scoped>\n", |
|
|
84 |
" .dataframe tbody tr th:only-of-type {\n", |
|
|
85 |
" vertical-align: middle;\n", |
|
|
86 |
" }\n", |
|
|
87 |
"\n", |
|
|
88 |
" .dataframe tbody tr th {\n", |
|
|
89 |
" vertical-align: top;\n", |
|
|
90 |
" }\n", |
|
|
91 |
"\n", |
|
|
92 |
" .dataframe thead th {\n", |
|
|
93 |
" text-align: right;\n", |
|
|
94 |
" }\n", |
|
|
95 |
"</style>\n", |
|
|
96 |
"<table border=\"1\" class=\"dataframe\">\n", |
|
|
97 |
" <thead>\n", |
|
|
98 |
" <tr style=\"text-align: right;\">\n", |
|
|
99 |
" <th></th>\n", |
|
|
100 |
" <th>ID</th>\n", |
|
|
101 |
" <th>Label</th>\n", |
|
|
102 |
" <th>Sub_type</th>\n", |
|
|
103 |
" <th>PatientID</th>\n", |
|
|
104 |
" </tr>\n", |
|
|
105 |
" </thead>\n", |
|
|
106 |
" <tbody>\n", |
|
|
107 |
" <tr>\n", |
|
|
108 |
" <th>0</th>\n", |
|
|
109 |
" <td>ID_28fbab7eb_epidural</td>\n", |
|
|
110 |
" <td>0.5</td>\n", |
|
|
111 |
" <td>epidural</td>\n", |
|
|
112 |
" <td>28fbab7eb</td>\n", |
|
|
113 |
" </tr>\n", |
|
|
114 |
" <tr>\n", |
|
|
115 |
" <th>1</th>\n", |
|
|
116 |
" <td>ID_28fbab7eb_intraparenchymal</td>\n", |
|
|
117 |
" <td>0.5</td>\n", |
|
|
118 |
" <td>intraparenchymal</td>\n", |
|
|
119 |
" <td>28fbab7eb</td>\n", |
|
|
120 |
" </tr>\n", |
|
|
121 |
" <tr>\n", |
|
|
122 |
" <th>2</th>\n", |
|
|
123 |
" <td>ID_28fbab7eb_intraventricular</td>\n", |
|
|
124 |
" <td>0.5</td>\n", |
|
|
125 |
" <td>intraventricular</td>\n", |
|
|
126 |
" <td>28fbab7eb</td>\n", |
|
|
127 |
" </tr>\n", |
|
|
128 |
" <tr>\n", |
|
|
129 |
" <th>3</th>\n", |
|
|
130 |
" <td>ID_28fbab7eb_subarachnoid</td>\n", |
|
|
131 |
" <td>0.5</td>\n", |
|
|
132 |
" <td>subarachnoid</td>\n", |
|
|
133 |
" <td>28fbab7eb</td>\n", |
|
|
134 |
" </tr>\n", |
|
|
135 |
" <tr>\n", |
|
|
136 |
" <th>4</th>\n", |
|
|
137 |
" <td>ID_28fbab7eb_subdural</td>\n", |
|
|
138 |
" <td>0.5</td>\n", |
|
|
139 |
" <td>subdural</td>\n", |
|
|
140 |
" <td>28fbab7eb</td>\n", |
|
|
141 |
" </tr>\n", |
|
|
142 |
" </tbody>\n", |
|
|
143 |
"</table>\n", |
|
|
144 |
"</div>" |
|
|
145 |
], |
|
|
146 |
"text/plain": [ |
|
|
147 |
" ID Label Sub_type PatientID\n", |
|
|
148 |
"0 ID_28fbab7eb_epidural 0.5 epidural 28fbab7eb\n", |
|
|
149 |
"1 ID_28fbab7eb_intraparenchymal 0.5 intraparenchymal 28fbab7eb\n", |
|
|
150 |
"2 ID_28fbab7eb_intraventricular 0.5 intraventricular 28fbab7eb\n", |
|
|
151 |
"3 ID_28fbab7eb_subarachnoid 0.5 subarachnoid 28fbab7eb\n", |
|
|
152 |
"4 ID_28fbab7eb_subdural 0.5 subdural 28fbab7eb" |
|
|
153 |
] |
|
|
154 |
}, |
|
|
155 |
"execution_count": 4, |
|
|
156 |
"metadata": {}, |
|
|
157 |
"output_type": "execute_result" |
|
|
158 |
} |
|
|
159 |
], |
|
|
160 |
"source": [ |
|
|
161 |
"train_base_df['Sub_type'] = train_base_df['ID'].str.split(\"_\", n = 3, expand = True)[2]\n", |
|
|
162 |
"train_base_df['PatientID'] = train_base_df['ID'].str.split(\"_\", n = 3, expand = True)[1]\n", |
|
|
163 |
"train_base_df.head()" |
|
|
164 |
] |
|
|
165 |
}, |
|
|
166 |
{ |
|
|
167 |
"cell_type": "code", |
|
|
168 |
"execution_count": 26, |
|
|
169 |
"metadata": {}, |
|
|
170 |
"outputs": [ |
|
|
171 |
{ |
|
|
172 |
"data": { |
|
|
173 |
"text/plain": [ |
|
|
174 |
"array(['epidural', 'intraparenchymal', 'intraventricular', 'subarachnoid',\n", |
|
|
175 |
" 'subdural', 'any'], dtype=object)" |
|
|
176 |
] |
|
|
177 |
}, |
|
|
178 |
"execution_count": 26, |
|
|
179 |
"metadata": {}, |
|
|
180 |
"output_type": "execute_result" |
|
|
181 |
} |
|
|
182 |
], |
|
|
183 |
"source": [ |
|
|
184 |
"sub_types=train_base_df.Sub_type.unique()\n", |
|
|
185 |
"sub_types" |
|
|
186 |
] |
|
|
187 |
}, |
|
|
188 |
{ |
|
|
189 |
"cell_type": "code", |
|
|
190 |
"execution_count": 27, |
|
|
191 |
"metadata": {}, |
|
|
192 |
"outputs": [ |
|
|
193 |
{ |
|
|
194 |
"data": { |
|
|
195 |
"application/vnd.jupyter.widget-view+json": { |
|
|
196 |
"model_id": "ae4afd5b0d824567b34dd59b1b81e430", |
|
|
197 |
"version_major": 2, |
|
|
198 |
"version_minor": 0 |
|
|
199 |
}, |
|
|
200 |
"text/plain": [ |
|
|
201 |
"HBox(children=(IntProgress(value=0, max=6), HTML(value='')))" |
|
|
202 |
] |
|
|
203 |
}, |
|
|
204 |
"metadata": {}, |
|
|
205 |
"output_type": "display_data" |
|
|
206 |
}, |
|
|
207 |
{ |
|
|
208 |
"name": "stdout", |
|
|
209 |
"output_type": "stream", |
|
|
210 |
"text": [ |
|
|
211 |
"\n" |
|
|
212 |
] |
|
|
213 |
} |
|
|
214 |
], |
|
|
215 |
"source": [ |
|
|
216 |
"dfs =[]\n", |
|
|
217 |
"for sub_type in tqdm_notebook(sub_types):\n", |
|
|
218 |
" df = train_base_df[train_base_df.Sub_type==sub_type][['PatientID','Label']].copy()\n", |
|
|
219 |
" df=df.rename(columns={\"Label\": sub_type}).reset_index(drop=True)\n", |
|
|
220 |
" dfs.append(df)\n", |
|
|
221 |
"train_df=dfs[0]\n", |
|
|
222 |
"for df in tqdm_notebook(dfs[1:]):\n", |
|
|
223 |
" train_df=test_df.merge(df,on='PatientID')" |
|
|
224 |
] |
|
|
225 |
}, |
|
|
226 |
{ |
|
|
227 |
"cell_type": "code", |
|
|
228 |
"execution_count": 11, |
|
|
229 |
"metadata": {}, |
|
|
230 |
"outputs": [ |
|
|
231 |
{ |
|
|
232 |
"data": { |
|
|
233 |
"application/vnd.jupyter.widget-view+json": { |
|
|
234 |
"model_id": "5fe21e059b0f4fff9decf2a356119816", |
|
|
235 |
"version_major": 2, |
|
|
236 |
"version_minor": 0 |
|
|
237 |
}, |
|
|
238 |
"text/plain": [ |
|
|
239 |
"HBox(children=(IntProgress(value=0, max=78545), HTML(value='')))" |
|
|
240 |
] |
|
|
241 |
}, |
|
|
242 |
"metadata": {}, |
|
|
243 |
"output_type": "display_data" |
|
|
244 |
}, |
|
|
245 |
{ |
|
|
246 |
"name": "stdout", |
|
|
247 |
"output_type": "stream", |
|
|
248 |
"text": [ |
|
|
249 |
"\n" |
|
|
250 |
] |
|
|
251 |
} |
|
|
252 |
], |
|
|
253 |
"source": [ |
|
|
254 |
"PID = np.zeros(train_df.shape[0],dtype=object)\n", |
|
|
255 |
"StudyI = np.zeros(train_df.shape[0],dtype=object)\n", |
|
|
256 |
"SeriesI = np.zeros(train_df.shape[0],dtype=object)\n", |
|
|
257 |
"WindowCenter = np.zeros(train_df.shape[0],dtype=object)\n", |
|
|
258 |
"WindowWidth = np.zeros(train_df.shape[0],dtype=object)\n", |
|
|
259 |
"ImagePositionX = np.zeros(train_df.shape[0],dtype=np.float)\n", |
|
|
260 |
"ImagePositionY = np.zeros(train_df.shape[0],dtype=np.float)\n", |
|
|
261 |
"ImagePositionZ = np.zeros(train_df.shape[0],dtype=np.float)\n", |
|
|
262 |
"\n", |
|
|
263 |
"for i,row in tqdm_notebook(train_df.iterrows(),total=train_df.shape[0]):\n", |
|
|
264 |
" ds = pydicom.dcmread(train_images_dir + 'ID_{}.dcm'.format(row['PatientID']))\n", |
|
|
265 |
" SeriesI[i]=ds.SeriesInstanceUID\n", |
|
|
266 |
" PID[i]=ds.PatientID\n", |
|
|
267 |
" StudyI[i]=ds.StudyInstanceUID \n", |
|
|
268 |
" WindowCenter[i]=ds.WindowCenter\n", |
|
|
269 |
" WindowWidth[i]=ds.WindowWidth\n", |
|
|
270 |
" ImagePositionX[i]=float(ds.ImagePositionPatient[0])\n", |
|
|
271 |
" ImagePositionY[i]=float(ds.ImagePositionPatient[1])\n", |
|
|
272 |
" ImagePositionZ[i]=float(ds.ImagePositionPatient[2])\n", |
|
|
273 |
"train_df['SeriesI']=SeriesI\n", |
|
|
274 |
"train_df['PID']=PID\n", |
|
|
275 |
"train_df['StudyI']=StudyI\n", |
|
|
276 |
"train_df['WindowCenter']=WindowCenter\n", |
|
|
277 |
"train_df['WindowWidth']=WindowWidth\n", |
|
|
278 |
"train_df['ImagePositionZ']=ImagePositionZ\n", |
|
|
279 |
"train_df['ImagePositionX']=ImagePositionX\n", |
|
|
280 |
"train_df['ImagePositionY']=ImagePositionY\n" |
|
|
281 |
] |
|
|
282 |
}, |
|
|
283 |
{ |
|
|
284 |
"cell_type": "code", |
|
|
285 |
"execution_count": null, |
|
|
286 |
"metadata": {}, |
|
|
287 |
"outputs": [], |
|
|
288 |
"source": [ |
|
|
289 |
"train_df.to_csv(data_dir+'train.csv',index=False)" |
|
|
290 |
] |
|
|
291 |
}, |
|
|
292 |
{ |
|
|
293 |
"cell_type": "code", |
|
|
294 |
"execution_count": 3, |
|
|
295 |
"metadata": {}, |
|
|
296 |
"outputs": [ |
|
|
297 |
{ |
|
|
298 |
"data": { |
|
|
299 |
"text/html": [ |
|
|
300 |
"<div>\n", |
|
|
301 |
"<style scoped>\n", |
|
|
302 |
" .dataframe tbody tr th:only-of-type {\n", |
|
|
303 |
" vertical-align: middle;\n", |
|
|
304 |
" }\n", |
|
|
305 |
"\n", |
|
|
306 |
" .dataframe tbody tr th {\n", |
|
|
307 |
" vertical-align: top;\n", |
|
|
308 |
" }\n", |
|
|
309 |
"\n", |
|
|
310 |
" .dataframe thead th {\n", |
|
|
311 |
" text-align: right;\n", |
|
|
312 |
" }\n", |
|
|
313 |
"</style>\n", |
|
|
314 |
"<table border=\"1\" class=\"dataframe\">\n", |
|
|
315 |
" <thead>\n", |
|
|
316 |
" <tr style=\"text-align: right;\">\n", |
|
|
317 |
" <th></th>\n", |
|
|
318 |
" <th>ID</th>\n", |
|
|
319 |
" <th>Label</th>\n", |
|
|
320 |
" </tr>\n", |
|
|
321 |
" </thead>\n", |
|
|
322 |
" <tbody>\n", |
|
|
323 |
" <tr>\n", |
|
|
324 |
" <th>0</th>\n", |
|
|
325 |
" <td>ID_28fbab7eb_epidural</td>\n", |
|
|
326 |
" <td>0.5</td>\n", |
|
|
327 |
" </tr>\n", |
|
|
328 |
" <tr>\n", |
|
|
329 |
" <th>1</th>\n", |
|
|
330 |
" <td>ID_28fbab7eb_intraparenchymal</td>\n", |
|
|
331 |
" <td>0.5</td>\n", |
|
|
332 |
" </tr>\n", |
|
|
333 |
" <tr>\n", |
|
|
334 |
" <th>2</th>\n", |
|
|
335 |
" <td>ID_28fbab7eb_intraventricular</td>\n", |
|
|
336 |
" <td>0.5</td>\n", |
|
|
337 |
" </tr>\n", |
|
|
338 |
" <tr>\n", |
|
|
339 |
" <th>3</th>\n", |
|
|
340 |
" <td>ID_28fbab7eb_subarachnoid</td>\n", |
|
|
341 |
" <td>0.5</td>\n", |
|
|
342 |
" </tr>\n", |
|
|
343 |
" <tr>\n", |
|
|
344 |
" <th>4</th>\n", |
|
|
345 |
" <td>ID_28fbab7eb_subdural</td>\n", |
|
|
346 |
" <td>0.5</td>\n", |
|
|
347 |
" </tr>\n", |
|
|
348 |
" </tbody>\n", |
|
|
349 |
"</table>\n", |
|
|
350 |
"</div>" |
|
|
351 |
], |
|
|
352 |
"text/plain": [ |
|
|
353 |
" ID Label\n", |
|
|
354 |
"0 ID_28fbab7eb_epidural 0.5\n", |
|
|
355 |
"1 ID_28fbab7eb_intraparenchymal 0.5\n", |
|
|
356 |
"2 ID_28fbab7eb_intraventricular 0.5\n", |
|
|
357 |
"3 ID_28fbab7eb_subarachnoid 0.5\n", |
|
|
358 |
"4 ID_28fbab7eb_subdural 0.5" |
|
|
359 |
] |
|
|
360 |
}, |
|
|
361 |
"execution_count": 3, |
|
|
362 |
"metadata": {}, |
|
|
363 |
"output_type": "execute_result" |
|
|
364 |
} |
|
|
365 |
], |
|
|
366 |
"source": [ |
|
|
367 |
"sample_submission=pd.read_csv(data_dir+'stage_1_sample_submission.csv')\n", |
|
|
368 |
"sample_submission.head()" |
|
|
369 |
] |
|
|
370 |
}, |
|
|
371 |
{ |
|
|
372 |
"cell_type": "code", |
|
|
373 |
"execution_count": 4, |
|
|
374 |
"metadata": {}, |
|
|
375 |
"outputs": [ |
|
|
376 |
{ |
|
|
377 |
"data": { |
|
|
378 |
"text/html": [ |
|
|
379 |
"<div>\n", |
|
|
380 |
"<style scoped>\n", |
|
|
381 |
" .dataframe tbody tr th:only-of-type {\n", |
|
|
382 |
" vertical-align: middle;\n", |
|
|
383 |
" }\n", |
|
|
384 |
"\n", |
|
|
385 |
" .dataframe tbody tr th {\n", |
|
|
386 |
" vertical-align: top;\n", |
|
|
387 |
" }\n", |
|
|
388 |
"\n", |
|
|
389 |
" .dataframe thead th {\n", |
|
|
390 |
" text-align: right;\n", |
|
|
391 |
" }\n", |
|
|
392 |
"</style>\n", |
|
|
393 |
"<table border=\"1\" class=\"dataframe\">\n", |
|
|
394 |
" <thead>\n", |
|
|
395 |
" <tr style=\"text-align: right;\">\n", |
|
|
396 |
" <th></th>\n", |
|
|
397 |
" <th>ID</th>\n", |
|
|
398 |
" <th>Label</th>\n", |
|
|
399 |
" <th>Sub_type</th>\n", |
|
|
400 |
" <th>PatientID</th>\n", |
|
|
401 |
" </tr>\n", |
|
|
402 |
" </thead>\n", |
|
|
403 |
" <tbody>\n", |
|
|
404 |
" <tr>\n", |
|
|
405 |
" <th>0</th>\n", |
|
|
406 |
" <td>ID_28fbab7eb_epidural</td>\n", |
|
|
407 |
" <td>0.5</td>\n", |
|
|
408 |
" <td>epidural</td>\n", |
|
|
409 |
" <td>28fbab7eb</td>\n", |
|
|
410 |
" </tr>\n", |
|
|
411 |
" <tr>\n", |
|
|
412 |
" <th>1</th>\n", |
|
|
413 |
" <td>ID_28fbab7eb_intraparenchymal</td>\n", |
|
|
414 |
" <td>0.5</td>\n", |
|
|
415 |
" <td>intraparenchymal</td>\n", |
|
|
416 |
" <td>28fbab7eb</td>\n", |
|
|
417 |
" </tr>\n", |
|
|
418 |
" <tr>\n", |
|
|
419 |
" <th>2</th>\n", |
|
|
420 |
" <td>ID_28fbab7eb_intraventricular</td>\n", |
|
|
421 |
" <td>0.5</td>\n", |
|
|
422 |
" <td>intraventricular</td>\n", |
|
|
423 |
" <td>28fbab7eb</td>\n", |
|
|
424 |
" </tr>\n", |
|
|
425 |
" <tr>\n", |
|
|
426 |
" <th>3</th>\n", |
|
|
427 |
" <td>ID_28fbab7eb_subarachnoid</td>\n", |
|
|
428 |
" <td>0.5</td>\n", |
|
|
429 |
" <td>subarachnoid</td>\n", |
|
|
430 |
" <td>28fbab7eb</td>\n", |
|
|
431 |
" </tr>\n", |
|
|
432 |
" <tr>\n", |
|
|
433 |
" <th>4</th>\n", |
|
|
434 |
" <td>ID_28fbab7eb_subdural</td>\n", |
|
|
435 |
" <td>0.5</td>\n", |
|
|
436 |
" <td>subdural</td>\n", |
|
|
437 |
" <td>28fbab7eb</td>\n", |
|
|
438 |
" </tr>\n", |
|
|
439 |
" </tbody>\n", |
|
|
440 |
"</table>\n", |
|
|
441 |
"</div>" |
|
|
442 |
], |
|
|
443 |
"text/plain": [ |
|
|
444 |
" ID Label Sub_type PatientID\n", |
|
|
445 |
"0 ID_28fbab7eb_epidural 0.5 epidural 28fbab7eb\n", |
|
|
446 |
"1 ID_28fbab7eb_intraparenchymal 0.5 intraparenchymal 28fbab7eb\n", |
|
|
447 |
"2 ID_28fbab7eb_intraventricular 0.5 intraventricular 28fbab7eb\n", |
|
|
448 |
"3 ID_28fbab7eb_subarachnoid 0.5 subarachnoid 28fbab7eb\n", |
|
|
449 |
"4 ID_28fbab7eb_subdural 0.5 subdural 28fbab7eb" |
|
|
450 |
] |
|
|
451 |
}, |
|
|
452 |
"execution_count": 4, |
|
|
453 |
"metadata": {}, |
|
|
454 |
"output_type": "execute_result" |
|
|
455 |
} |
|
|
456 |
], |
|
|
457 |
"source": [ |
|
|
458 |
"test_base_df=sample_submission.copy()\n", |
|
|
459 |
"test_base_df['Sub_type'] = test_base_df['ID'].str.split(\"_\", n = 3, expand = True)[2]\n", |
|
|
460 |
"test_base_df['PatientID'] = test_base_df['ID'].str.split(\"_\", n = 3, expand = True)[1]\n", |
|
|
461 |
"test_base_df.head()" |
|
|
462 |
] |
|
|
463 |
}, |
|
|
464 |
{ |
|
|
465 |
"cell_type": "code", |
|
|
466 |
"execution_count": 234, |
|
|
467 |
"metadata": {}, |
|
|
468 |
"outputs": [], |
|
|
469 |
"source": [ |
|
|
470 |
"test_ids=test_df.PatientID.unique()" |
|
|
471 |
] |
|
|
472 |
}, |
|
|
473 |
{ |
|
|
474 |
"cell_type": "code", |
|
|
475 |
"execution_count": 26, |
|
|
476 |
"metadata": {}, |
|
|
477 |
"outputs": [ |
|
|
478 |
{ |
|
|
479 |
"data": { |
|
|
480 |
"text/plain": [ |
|
|
481 |
"array(['epidural', 'intraparenchymal', 'intraventricular', 'subarachnoid',\n", |
|
|
482 |
" 'subdural', 'any'], dtype=object)" |
|
|
483 |
] |
|
|
484 |
}, |
|
|
485 |
"execution_count": 26, |
|
|
486 |
"metadata": {}, |
|
|
487 |
"output_type": "execute_result" |
|
|
488 |
} |
|
|
489 |
], |
|
|
490 |
"source": [ |
|
|
491 |
"sub_types=test_base_df.Sub_type.unique()\n", |
|
|
492 |
"sub_types" |
|
|
493 |
] |
|
|
494 |
}, |
|
|
495 |
{ |
|
|
496 |
"cell_type": "code", |
|
|
497 |
"execution_count": 7, |
|
|
498 |
"metadata": {}, |
|
|
499 |
"outputs": [ |
|
|
500 |
{ |
|
|
501 |
"data": { |
|
|
502 |
"application/vnd.jupyter.widget-view+json": { |
|
|
503 |
"model_id": "f153620a2b61441bb7cb7f1acf392844", |
|
|
504 |
"version_major": 2, |
|
|
505 |
"version_minor": 0 |
|
|
506 |
}, |
|
|
507 |
"text/plain": [ |
|
|
508 |
"HBox(children=(IntProgress(value=0, max=6), HTML(value='')))" |
|
|
509 |
] |
|
|
510 |
}, |
|
|
511 |
"metadata": {}, |
|
|
512 |
"output_type": "display_data" |
|
|
513 |
}, |
|
|
514 |
{ |
|
|
515 |
"name": "stdout", |
|
|
516 |
"output_type": "stream", |
|
|
517 |
"text": [ |
|
|
518 |
"\n" |
|
|
519 |
] |
|
|
520 |
}, |
|
|
521 |
{ |
|
|
522 |
"data": { |
|
|
523 |
"application/vnd.jupyter.widget-view+json": { |
|
|
524 |
"model_id": "189e8ac701514c2c963161608d43a459", |
|
|
525 |
"version_major": 2, |
|
|
526 |
"version_minor": 0 |
|
|
527 |
}, |
|
|
528 |
"text/plain": [ |
|
|
529 |
"HBox(children=(IntProgress(value=0, max=5), HTML(value='')))" |
|
|
530 |
] |
|
|
531 |
}, |
|
|
532 |
"metadata": {}, |
|
|
533 |
"output_type": "display_data" |
|
|
534 |
}, |
|
|
535 |
{ |
|
|
536 |
"name": "stdout", |
|
|
537 |
"output_type": "stream", |
|
|
538 |
"text": [ |
|
|
539 |
"\n" |
|
|
540 |
] |
|
|
541 |
} |
|
|
542 |
], |
|
|
543 |
"source": [ |
|
|
544 |
"dfs =[]\n", |
|
|
545 |
"for sub_type in tqdm_notebook(sub_types):\n", |
|
|
546 |
" df = test_base_df[test_base_df.Sub_type==sub_type][['PatientID','Label']].copy()\n", |
|
|
547 |
" df=df.rename(columns={\"Label\": sub_type}).reset_index(drop=True)\n", |
|
|
548 |
" dfs.append(df)\n", |
|
|
549 |
"test_df=dfs[0]\n", |
|
|
550 |
"for df in tqdm_notebook(dfs[1:]):\n", |
|
|
551 |
" test_df=test_df.merge(df,on='PatientID')" |
|
|
552 |
] |
|
|
553 |
}, |
|
|
554 |
{ |
|
|
555 |
"cell_type": "code", |
|
|
556 |
"execution_count": 8, |
|
|
557 |
"metadata": {}, |
|
|
558 |
"outputs": [ |
|
|
559 |
{ |
|
|
560 |
"data": { |
|
|
561 |
"text/html": [ |
|
|
562 |
"<div>\n", |
|
|
563 |
"<style scoped>\n", |
|
|
564 |
" .dataframe tbody tr th:only-of-type {\n", |
|
|
565 |
" vertical-align: middle;\n", |
|
|
566 |
" }\n", |
|
|
567 |
"\n", |
|
|
568 |
" .dataframe tbody tr th {\n", |
|
|
569 |
" vertical-align: top;\n", |
|
|
570 |
" }\n", |
|
|
571 |
"\n", |
|
|
572 |
" .dataframe thead th {\n", |
|
|
573 |
" text-align: right;\n", |
|
|
574 |
" }\n", |
|
|
575 |
"</style>\n", |
|
|
576 |
"<table border=\"1\" class=\"dataframe\">\n", |
|
|
577 |
" <thead>\n", |
|
|
578 |
" <tr style=\"text-align: right;\">\n", |
|
|
579 |
" <th></th>\n", |
|
|
580 |
" <th>PatientID</th>\n", |
|
|
581 |
" <th>epidural</th>\n", |
|
|
582 |
" <th>intraparenchymal</th>\n", |
|
|
583 |
" <th>intraventricular</th>\n", |
|
|
584 |
" <th>subarachnoid</th>\n", |
|
|
585 |
" <th>subdural</th>\n", |
|
|
586 |
" <th>any</th>\n", |
|
|
587 |
" </tr>\n", |
|
|
588 |
" </thead>\n", |
|
|
589 |
" <tbody>\n", |
|
|
590 |
" <tr>\n", |
|
|
591 |
" <th>0</th>\n", |
|
|
592 |
" <td>28fbab7eb</td>\n", |
|
|
593 |
" <td>0.5</td>\n", |
|
|
594 |
" <td>0.5</td>\n", |
|
|
595 |
" <td>0.5</td>\n", |
|
|
596 |
" <td>0.5</td>\n", |
|
|
597 |
" <td>0.5</td>\n", |
|
|
598 |
" <td>0.5</td>\n", |
|
|
599 |
" </tr>\n", |
|
|
600 |
" <tr>\n", |
|
|
601 |
" <th>1</th>\n", |
|
|
602 |
" <td>877923b8b</td>\n", |
|
|
603 |
" <td>0.5</td>\n", |
|
|
604 |
" <td>0.5</td>\n", |
|
|
605 |
" <td>0.5</td>\n", |
|
|
606 |
" <td>0.5</td>\n", |
|
|
607 |
" <td>0.5</td>\n", |
|
|
608 |
" <td>0.5</td>\n", |
|
|
609 |
" </tr>\n", |
|
|
610 |
" <tr>\n", |
|
|
611 |
" <th>2</th>\n", |
|
|
612 |
" <td>a591477cb</td>\n", |
|
|
613 |
" <td>0.5</td>\n", |
|
|
614 |
" <td>0.5</td>\n", |
|
|
615 |
" <td>0.5</td>\n", |
|
|
616 |
" <td>0.5</td>\n", |
|
|
617 |
" <td>0.5</td>\n", |
|
|
618 |
" <td>0.5</td>\n", |
|
|
619 |
" </tr>\n", |
|
|
620 |
" <tr>\n", |
|
|
621 |
" <th>3</th>\n", |
|
|
622 |
" <td>42217c898</td>\n", |
|
|
623 |
" <td>0.5</td>\n", |
|
|
624 |
" <td>0.5</td>\n", |
|
|
625 |
" <td>0.5</td>\n", |
|
|
626 |
" <td>0.5</td>\n", |
|
|
627 |
" <td>0.5</td>\n", |
|
|
628 |
" <td>0.5</td>\n", |
|
|
629 |
" </tr>\n", |
|
|
630 |
" <tr>\n", |
|
|
631 |
" <th>4</th>\n", |
|
|
632 |
" <td>a130c4d2f</td>\n", |
|
|
633 |
" <td>0.5</td>\n", |
|
|
634 |
" <td>0.5</td>\n", |
|
|
635 |
" <td>0.5</td>\n", |
|
|
636 |
" <td>0.5</td>\n", |
|
|
637 |
" <td>0.5</td>\n", |
|
|
638 |
" <td>0.5</td>\n", |
|
|
639 |
" </tr>\n", |
|
|
640 |
" </tbody>\n", |
|
|
641 |
"</table>\n", |
|
|
642 |
"</div>" |
|
|
643 |
], |
|
|
644 |
"text/plain": [ |
|
|
645 |
" PatientID epidural intraparenchymal intraventricular subarachnoid \\\n", |
|
|
646 |
"0 28fbab7eb 0.5 0.5 0.5 0.5 \n", |
|
|
647 |
"1 877923b8b 0.5 0.5 0.5 0.5 \n", |
|
|
648 |
"2 a591477cb 0.5 0.5 0.5 0.5 \n", |
|
|
649 |
"3 42217c898 0.5 0.5 0.5 0.5 \n", |
|
|
650 |
"4 a130c4d2f 0.5 0.5 0.5 0.5 \n", |
|
|
651 |
"\n", |
|
|
652 |
" subdural any \n", |
|
|
653 |
"0 0.5 0.5 \n", |
|
|
654 |
"1 0.5 0.5 \n", |
|
|
655 |
"2 0.5 0.5 \n", |
|
|
656 |
"3 0.5 0.5 \n", |
|
|
657 |
"4 0.5 0.5 " |
|
|
658 |
] |
|
|
659 |
}, |
|
|
660 |
"execution_count": 8, |
|
|
661 |
"metadata": {}, |
|
|
662 |
"output_type": "execute_result" |
|
|
663 |
} |
|
|
664 |
], |
|
|
665 |
"source": [ |
|
|
666 |
"test_df.head()" |
|
|
667 |
] |
|
|
668 |
}, |
|
|
669 |
{ |
|
|
670 |
"cell_type": "code", |
|
|
671 |
"execution_count": 11, |
|
|
672 |
"metadata": {}, |
|
|
673 |
"outputs": [ |
|
|
674 |
{ |
|
|
675 |
"data": { |
|
|
676 |
"application/vnd.jupyter.widget-view+json": { |
|
|
677 |
"model_id": "5fe21e059b0f4fff9decf2a356119816", |
|
|
678 |
"version_major": 2, |
|
|
679 |
"version_minor": 0 |
|
|
680 |
}, |
|
|
681 |
"text/plain": [ |
|
|
682 |
"HBox(children=(IntProgress(value=0, max=78545), HTML(value='')))" |
|
|
683 |
] |
|
|
684 |
}, |
|
|
685 |
"metadata": {}, |
|
|
686 |
"output_type": "display_data" |
|
|
687 |
}, |
|
|
688 |
{ |
|
|
689 |
"name": "stdout", |
|
|
690 |
"output_type": "stream", |
|
|
691 |
"text": [ |
|
|
692 |
"\n" |
|
|
693 |
] |
|
|
694 |
} |
|
|
695 |
], |
|
|
696 |
"source": [ |
|
|
697 |
"PID = np.zeros(test_df.shape[0],dtype=object)\n", |
|
|
698 |
"StudyI = np.zeros(test_df.shape[0],dtype=object)\n", |
|
|
699 |
"SeriesI = np.zeros(test_df.shape[0],dtype=object)\n", |
|
|
700 |
"WindowCenter = np.zeros(test_df.shape[0],dtype=object)\n", |
|
|
701 |
"WindowWidth = np.zeros(test_df.shape[0],dtype=object)\n", |
|
|
702 |
"ImagePositionX = np.zeros(test_df.shape[0],dtype=np.float)\n", |
|
|
703 |
"ImagePositionY = np.zeros(test_df.shape[0],dtype=np.float)\n", |
|
|
704 |
"ImagePositionZ = np.zeros(test_df.shape[0],dtype=np.float)\n", |
|
|
705 |
"\n", |
|
|
706 |
"for i,row in tqdm_notebook(test_df.iterrows(),total=test_df.shape[0]):\n", |
|
|
707 |
" ds = pydicom.dcmread(test_images_dir + 'ID_{}.dcm'.format(row['PatientID']))\n", |
|
|
708 |
" SeriesI[i]=ds.SeriesInstanceUID\n", |
|
|
709 |
" PID[i]=ds.PatientID\n", |
|
|
710 |
" StudyI[i]=ds.StudyInstanceUID \n", |
|
|
711 |
" WindowCenter[i]=ds.WindowCenter\n", |
|
|
712 |
" WindowWidth[i]=ds.WindowWidth\n", |
|
|
713 |
" ImagePositionX[i]=float(ds.ImagePositionPatient[0])\n", |
|
|
714 |
" ImagePositionY[i]=float(ds.ImagePositionPatient[1])\n", |
|
|
715 |
" ImagePositionZ[i]=float(ds.ImagePositionPatient[2])\n", |
|
|
716 |
"test_df['SeriesI']=SeriesI\n", |
|
|
717 |
"test_df['PID']=PID\n", |
|
|
718 |
"test_df['StudyI']=StudyI\n", |
|
|
719 |
"test_df['WindowCenter']=WindowCenter\n", |
|
|
720 |
"test_df['WindowWidth']=WindowWidth\n", |
|
|
721 |
"test_df['ImagePositionZ']=ImagePositionZ\n", |
|
|
722 |
"test_df['ImagePositionX']=ImagePositionX\n", |
|
|
723 |
"test_df['ImagePositionY']=ImagePositionY\n" |
|
|
724 |
] |
|
|
725 |
}, |
|
|
726 |
{ |
|
|
727 |
"cell_type": "code", |
|
|
728 |
"execution_count": 23, |
|
|
729 |
"metadata": {}, |
|
|
730 |
"outputs": [], |
|
|
731 |
"source": [ |
|
|
732 |
"test_df.to_csv(data_dir+'test.csv',index=False)" |
|
|
733 |
] |
|
|
734 |
}, |
|
|
735 |
{ |
|
|
736 |
"cell_type": "code", |
|
|
737 |
"execution_count": null, |
|
|
738 |
"metadata": {}, |
|
|
739 |
"outputs": [], |
|
|
740 |
"source": [] |
|
|
741 |
} |
|
|
742 |
], |
|
|
743 |
"metadata": { |
|
|
744 |
"kernelspec": { |
|
|
745 |
"display_name": "Python 3", |
|
|
746 |
"language": "python", |
|
|
747 |
"name": "python3" |
|
|
748 |
}, |
|
|
749 |
"language_info": { |
|
|
750 |
"codemirror_mode": { |
|
|
751 |
"name": "ipython", |
|
|
752 |
"version": 3 |
|
|
753 |
}, |
|
|
754 |
"file_extension": ".py", |
|
|
755 |
"mimetype": "text/x-python", |
|
|
756 |
"name": "python", |
|
|
757 |
"nbconvert_exporter": "python", |
|
|
758 |
"pygments_lexer": "ipython3", |
|
|
759 |
"version": "3.6.6" |
|
|
760 |
} |
|
|
761 |
}, |
|
|
762 |
"nbformat": 4, |
|
|
763 |
"nbformat_minor": 2 |
|
|
764 |
} |