Datasets
Models
Downloads: 1
Diff of
/0-preprocess-generate_csvs.ipynb
[000000]
..
[fc9ccf]
Switch to unified view
| a | b/0-preprocess-generate_csvs.ipynb | ||
|---|---|---|---|
| 1 | { |
||
| 2 | "cells": [ |
||
| 3 | { |
||
| 4 | "cell_type": "code", |
||
| 5 | "execution_count": 1, |
||
| 6 | "metadata": {}, |
||
| 7 | "outputs": [], |
||
| 8 | "source": [ |
||
| 9 | "import pandas as pd\n", |
||
| 10 | "from pathlib import Path\n", |
||
| 11 | "from tqdm import tqdm_notebook\n", |
||
| 12 | "import pydicom\n", |
||
| 13 | "import itertools\n", |
||
| 14 | "import numpy as np\n", |
||
| 15 | "from concurrent.futures import ProcessPoolExecutor" |
||
| 16 | ] |
||
| 17 | }, |
||
| 18 | { |
||
| 19 | "cell_type": "code", |
||
| 20 | "execution_count": 2, |
||
| 21 | "metadata": {}, |
||
| 22 | "outputs": [], |
||
| 23 | "source": [ |
||
| 24 | "%%capture\n", |
||
| 25 | "from tqdm import tqdm_notebook as tqdm\n", |
||
| 26 | "tqdm().pandas()" |
||
| 27 | ] |
||
| 28 | }, |
||
| 29 | { |
||
| 30 | "cell_type": "markdown", |
||
| 31 | "metadata": {}, |
||
| 32 | "source": [ |
||
| 33 | "# Read stage_X_train and split id/label" |
||
| 34 | ] |
||
| 35 | }, |
||
| 36 | { |
||
| 37 | "cell_type": "code", |
||
| 38 | "execution_count": 3, |
||
| 39 | "metadata": {}, |
||
| 40 | "outputs": [], |
||
| 41 | "source": [ |
||
| 42 | "stage = \"stage_2\"" |
||
| 43 | ] |
||
| 44 | }, |
||
| 45 | { |
||
| 46 | "cell_type": "code", |
||
| 47 | "execution_count": 4, |
||
| 48 | "metadata": {}, |
||
| 49 | "outputs": [], |
||
| 50 | "source": [ |
||
| 51 | "bad_dcm_fn = f'data/unzip/{stage}_train_images/ID_6431af929.dcm'" |
||
| 52 | ] |
||
| 53 | }, |
||
| 54 | { |
||
| 55 | "cell_type": "code", |
||
| 56 | "execution_count": 5, |
||
| 57 | "metadata": {}, |
||
| 58 | "outputs": [ |
||
| 59 | { |
||
| 60 | "name": "stdout", |
||
| 61 | "output_type": "stream", |
||
| 62 | "text": [ |
||
| 63 | "rm: cannot remove 'data/unzip/stage_2_train_images/ID_6431af929.dcm': No such file or directory\r\n" |
||
| 64 | ] |
||
| 65 | } |
||
| 66 | ], |
||
| 67 | "source": [ |
||
| 68 | "!rm {bad_dcm_fn}" |
||
| 69 | ] |
||
| 70 | }, |
||
| 71 | { |
||
| 72 | "cell_type": "code", |
||
| 73 | "execution_count": 6, |
||
| 74 | "metadata": {}, |
||
| 75 | "outputs": [], |
||
| 76 | "source": [ |
||
| 77 | "df_train = pd.read_csv(f'data/unzip/{stage}_train.csv')" |
||
| 78 | ] |
||
| 79 | }, |
||
| 80 | { |
||
| 81 | "cell_type": "code", |
||
| 82 | "execution_count": 7, |
||
| 83 | "metadata": {}, |
||
| 84 | "outputs": [ |
||
| 85 | { |
||
| 86 | "data": { |
||
| 87 | "text/html": [ |
||
| 88 | "<div>\n", |
||
| 89 | "<style scoped>\n", |
||
| 90 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 91 | " vertical-align: middle;\n", |
||
| 92 | " }\n", |
||
| 93 | "\n", |
||
| 94 | " .dataframe tbody tr th {\n", |
||
| 95 | " vertical-align: top;\n", |
||
| 96 | " }\n", |
||
| 97 | "\n", |
||
| 98 | " .dataframe thead th {\n", |
||
| 99 | " text-align: right;\n", |
||
| 100 | " }\n", |
||
| 101 | "</style>\n", |
||
| 102 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 103 | " <thead>\n", |
||
| 104 | " <tr style=\"text-align: right;\">\n", |
||
| 105 | " <th></th>\n", |
||
| 106 | " <th>ID</th>\n", |
||
| 107 | " <th>Label</th>\n", |
||
| 108 | " </tr>\n", |
||
| 109 | " </thead>\n", |
||
| 110 | " <tbody>\n", |
||
| 111 | " <tr>\n", |
||
| 112 | " <th>0</th>\n", |
||
| 113 | " <td>ID_12cadc6af_epidural</td>\n", |
||
| 114 | " <td>0</td>\n", |
||
| 115 | " </tr>\n", |
||
| 116 | " <tr>\n", |
||
| 117 | " <th>1</th>\n", |
||
| 118 | " <td>ID_12cadc6af_intraparenchymal</td>\n", |
||
| 119 | " <td>0</td>\n", |
||
| 120 | " </tr>\n", |
||
| 121 | " <tr>\n", |
||
| 122 | " <th>2</th>\n", |
||
| 123 | " <td>ID_12cadc6af_intraventricular</td>\n", |
||
| 124 | " <td>0</td>\n", |
||
| 125 | " </tr>\n", |
||
| 126 | " <tr>\n", |
||
| 127 | " <th>3</th>\n", |
||
| 128 | " <td>ID_12cadc6af_subarachnoid</td>\n", |
||
| 129 | " <td>0</td>\n", |
||
| 130 | " </tr>\n", |
||
| 131 | " <tr>\n", |
||
| 132 | " <th>4</th>\n", |
||
| 133 | " <td>ID_12cadc6af_subdural</td>\n", |
||
| 134 | " <td>0</td>\n", |
||
| 135 | " </tr>\n", |
||
| 136 | " </tbody>\n", |
||
| 137 | "</table>\n", |
||
| 138 | "</div>" |
||
| 139 | ], |
||
| 140 | "text/plain": [ |
||
| 141 | " ID Label\n", |
||
| 142 | "0 ID_12cadc6af_epidural 0\n", |
||
| 143 | "1 ID_12cadc6af_intraparenchymal 0\n", |
||
| 144 | "2 ID_12cadc6af_intraventricular 0\n", |
||
| 145 | "3 ID_12cadc6af_subarachnoid 0\n", |
||
| 146 | "4 ID_12cadc6af_subdural 0" |
||
| 147 | ] |
||
| 148 | }, |
||
| 149 | "execution_count": 7, |
||
| 150 | "metadata": {}, |
||
| 151 | "output_type": "execute_result" |
||
| 152 | } |
||
| 153 | ], |
||
| 154 | "source": [ |
||
| 155 | "df_train.head()" |
||
| 156 | ] |
||
| 157 | }, |
||
| 158 | { |
||
| 159 | "cell_type": "code", |
||
| 160 | "execution_count": 8, |
||
| 161 | "metadata": {}, |
||
| 162 | "outputs": [], |
||
| 163 | "source": [ |
||
| 164 | "df_train['fid'] = df_train.ID.apply(lambda x: '_'.join(x.split('_')[:2]) )" |
||
| 165 | ] |
||
| 166 | }, |
||
| 167 | { |
||
| 168 | "cell_type": "code", |
||
| 169 | "execution_count": 9, |
||
| 170 | "metadata": {}, |
||
| 171 | "outputs": [], |
||
| 172 | "source": [ |
||
| 173 | "df_train.columns = ['ID', 'probability', 'fid']" |
||
| 174 | ] |
||
| 175 | }, |
||
| 176 | { |
||
| 177 | "cell_type": "code", |
||
| 178 | "execution_count": 10, |
||
| 179 | "metadata": {}, |
||
| 180 | "outputs": [], |
||
| 181 | "source": [ |
||
| 182 | "df_train['label'] = df_train.ID.apply(lambda x: x.split('_')[-1])" |
||
| 183 | ] |
||
| 184 | }, |
||
| 185 | { |
||
| 186 | "cell_type": "code", |
||
| 187 | "execution_count": 11, |
||
| 188 | "metadata": {}, |
||
| 189 | "outputs": [ |
||
| 190 | { |
||
| 191 | "data": { |
||
| 192 | "text/html": [ |
||
| 193 | "<div>\n", |
||
| 194 | "<style scoped>\n", |
||
| 195 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 196 | " vertical-align: middle;\n", |
||
| 197 | " }\n", |
||
| 198 | "\n", |
||
| 199 | " .dataframe tbody tr th {\n", |
||
| 200 | " vertical-align: top;\n", |
||
| 201 | " }\n", |
||
| 202 | "\n", |
||
| 203 | " .dataframe thead th {\n", |
||
| 204 | " text-align: right;\n", |
||
| 205 | " }\n", |
||
| 206 | "</style>\n", |
||
| 207 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 208 | " <thead>\n", |
||
| 209 | " <tr style=\"text-align: right;\">\n", |
||
| 210 | " <th></th>\n", |
||
| 211 | " <th>ID</th>\n", |
||
| 212 | " <th>probability</th>\n", |
||
| 213 | " <th>fid</th>\n", |
||
| 214 | " <th>label</th>\n", |
||
| 215 | " </tr>\n", |
||
| 216 | " </thead>\n", |
||
| 217 | " <tbody>\n", |
||
| 218 | " <tr>\n", |
||
| 219 | " <th>0</th>\n", |
||
| 220 | " <td>ID_12cadc6af_epidural</td>\n", |
||
| 221 | " <td>0</td>\n", |
||
| 222 | " <td>ID_12cadc6af</td>\n", |
||
| 223 | " <td>epidural</td>\n", |
||
| 224 | " </tr>\n", |
||
| 225 | " <tr>\n", |
||
| 226 | " <th>1</th>\n", |
||
| 227 | " <td>ID_12cadc6af_intraparenchymal</td>\n", |
||
| 228 | " <td>0</td>\n", |
||
| 229 | " <td>ID_12cadc6af</td>\n", |
||
| 230 | " <td>intraparenchymal</td>\n", |
||
| 231 | " </tr>\n", |
||
| 232 | " <tr>\n", |
||
| 233 | " <th>2</th>\n", |
||
| 234 | " <td>ID_12cadc6af_intraventricular</td>\n", |
||
| 235 | " <td>0</td>\n", |
||
| 236 | " <td>ID_12cadc6af</td>\n", |
||
| 237 | " <td>intraventricular</td>\n", |
||
| 238 | " </tr>\n", |
||
| 239 | " <tr>\n", |
||
| 240 | " <th>3</th>\n", |
||
| 241 | " <td>ID_12cadc6af_subarachnoid</td>\n", |
||
| 242 | " <td>0</td>\n", |
||
| 243 | " <td>ID_12cadc6af</td>\n", |
||
| 244 | " <td>subarachnoid</td>\n", |
||
| 245 | " </tr>\n", |
||
| 246 | " <tr>\n", |
||
| 247 | " <th>4</th>\n", |
||
| 248 | " <td>ID_12cadc6af_subdural</td>\n", |
||
| 249 | " <td>0</td>\n", |
||
| 250 | " <td>ID_12cadc6af</td>\n", |
||
| 251 | " <td>subdural</td>\n", |
||
| 252 | " </tr>\n", |
||
| 253 | " </tbody>\n", |
||
| 254 | "</table>\n", |
||
| 255 | "</div>" |
||
| 256 | ], |
||
| 257 | "text/plain": [ |
||
| 258 | " ID probability fid label\n", |
||
| 259 | "0 ID_12cadc6af_epidural 0 ID_12cadc6af epidural\n", |
||
| 260 | "1 ID_12cadc6af_intraparenchymal 0 ID_12cadc6af intraparenchymal\n", |
||
| 261 | "2 ID_12cadc6af_intraventricular 0 ID_12cadc6af intraventricular\n", |
||
| 262 | "3 ID_12cadc6af_subarachnoid 0 ID_12cadc6af subarachnoid\n", |
||
| 263 | "4 ID_12cadc6af_subdural 0 ID_12cadc6af subdural" |
||
| 264 | ] |
||
| 265 | }, |
||
| 266 | "execution_count": 11, |
||
| 267 | "metadata": {}, |
||
| 268 | "output_type": "execute_result" |
||
| 269 | } |
||
| 270 | ], |
||
| 271 | "source": [ |
||
| 272 | "df_train.head()" |
||
| 273 | ] |
||
| 274 | }, |
||
| 275 | { |
||
| 276 | "cell_type": "markdown", |
||
| 277 | "metadata": {}, |
||
| 278 | "source": [ |
||
| 279 | "# Remove dupes " |
||
| 280 | ] |
||
| 281 | }, |
||
| 282 | { |
||
| 283 | "cell_type": "code", |
||
| 284 | "execution_count": 12, |
||
| 285 | "metadata": {}, |
||
| 286 | "outputs": [ |
||
| 287 | { |
||
| 288 | "data": { |
||
| 289 | "text/plain": [ |
||
| 290 | "(4516842, 4)" |
||
| 291 | ] |
||
| 292 | }, |
||
| 293 | "execution_count": 12, |
||
| 294 | "metadata": {}, |
||
| 295 | "output_type": "execute_result" |
||
| 296 | } |
||
| 297 | ], |
||
| 298 | "source": [ |
||
| 299 | "df_train.shape" |
||
| 300 | ] |
||
| 301 | }, |
||
| 302 | { |
||
| 303 | "cell_type": "code", |
||
| 304 | "execution_count": 13, |
||
| 305 | "metadata": {}, |
||
| 306 | "outputs": [], |
||
| 307 | "source": [ |
||
| 308 | "df_train.drop_duplicates('ID', inplace=True)" |
||
| 309 | ] |
||
| 310 | }, |
||
| 311 | { |
||
| 312 | "cell_type": "code", |
||
| 313 | "execution_count": 14, |
||
| 314 | "metadata": {}, |
||
| 315 | "outputs": [ |
||
| 316 | { |
||
| 317 | "data": { |
||
| 318 | "text/plain": [ |
||
| 319 | "(4516818, 4)" |
||
| 320 | ] |
||
| 321 | }, |
||
| 322 | "execution_count": 14, |
||
| 323 | "metadata": {}, |
||
| 324 | "output_type": "execute_result" |
||
| 325 | } |
||
| 326 | ], |
||
| 327 | "source": [ |
||
| 328 | "df_train.shape" |
||
| 329 | ] |
||
| 330 | }, |
||
| 331 | { |
||
| 332 | "cell_type": "markdown", |
||
| 333 | "metadata": {}, |
||
| 334 | "source": [ |
||
| 335 | "# Remove corrupted image" |
||
| 336 | ] |
||
| 337 | }, |
||
| 338 | { |
||
| 339 | "cell_type": "code", |
||
| 340 | "execution_count": 15, |
||
| 341 | "metadata": {}, |
||
| 342 | "outputs": [], |
||
| 343 | "source": [ |
||
| 344 | "df_train = df_train[df_train.fid != 'ID_6431af929'] # ID_6431af929" |
||
| 345 | ] |
||
| 346 | }, |
||
| 347 | { |
||
| 348 | "cell_type": "code", |
||
| 349 | "execution_count": 16, |
||
| 350 | "metadata": {}, |
||
| 351 | "outputs": [ |
||
| 352 | { |
||
| 353 | "data": { |
||
| 354 | "text/plain": [ |
||
| 355 | "(4516812, 4)" |
||
| 356 | ] |
||
| 357 | }, |
||
| 358 | "execution_count": 16, |
||
| 359 | "metadata": {}, |
||
| 360 | "output_type": "execute_result" |
||
| 361 | } |
||
| 362 | ], |
||
| 363 | "source": [ |
||
| 364 | "df_train.shape" |
||
| 365 | ] |
||
| 366 | }, |
||
| 367 | { |
||
| 368 | "cell_type": "markdown", |
||
| 369 | "metadata": {}, |
||
| 370 | "source": [ |
||
| 371 | "# Create pivot table with diagnostic labels as columns\n", |
||
| 372 | "Generates:\n", |
||
| 373 | "* `train_diags.csv` (previously named `train_pivot.csv`)" |
||
| 374 | ] |
||
| 375 | }, |
||
| 376 | { |
||
| 377 | "cell_type": "code", |
||
| 378 | "execution_count": 17, |
||
| 379 | "metadata": {}, |
||
| 380 | "outputs": [], |
||
| 381 | "source": [ |
||
| 382 | "df_diags = df_train.pivot(index='fid', columns='label', values='probability')" |
||
| 383 | ] |
||
| 384 | }, |
||
| 385 | { |
||
| 386 | "cell_type": "code", |
||
| 387 | "execution_count": 18, |
||
| 388 | "metadata": {}, |
||
| 389 | "outputs": [ |
||
| 390 | { |
||
| 391 | "data": { |
||
| 392 | "text/html": [ |
||
| 393 | "<div>\n", |
||
| 394 | "<style scoped>\n", |
||
| 395 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 396 | " vertical-align: middle;\n", |
||
| 397 | " }\n", |
||
| 398 | "\n", |
||
| 399 | " .dataframe tbody tr th {\n", |
||
| 400 | " vertical-align: top;\n", |
||
| 401 | " }\n", |
||
| 402 | "\n", |
||
| 403 | " .dataframe thead th {\n", |
||
| 404 | " text-align: right;\n", |
||
| 405 | " }\n", |
||
| 406 | "</style>\n", |
||
| 407 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 408 | " <thead>\n", |
||
| 409 | " <tr style=\"text-align: right;\">\n", |
||
| 410 | " <th>label</th>\n", |
||
| 411 | " <th>any</th>\n", |
||
| 412 | " <th>epidural</th>\n", |
||
| 413 | " <th>intraparenchymal</th>\n", |
||
| 414 | " <th>intraventricular</th>\n", |
||
| 415 | " <th>subarachnoid</th>\n", |
||
| 416 | " <th>subdural</th>\n", |
||
| 417 | " </tr>\n", |
||
| 418 | " <tr>\n", |
||
| 419 | " <th>fid</th>\n", |
||
| 420 | " <th></th>\n", |
||
| 421 | " <th></th>\n", |
||
| 422 | " <th></th>\n", |
||
| 423 | " <th></th>\n", |
||
| 424 | " <th></th>\n", |
||
| 425 | " <th></th>\n", |
||
| 426 | " </tr>\n", |
||
| 427 | " </thead>\n", |
||
| 428 | " <tbody>\n", |
||
| 429 | " <tr>\n", |
||
| 430 | " <th>ID_000012eaf</th>\n", |
||
| 431 | " <td>0</td>\n", |
||
| 432 | " <td>0</td>\n", |
||
| 433 | " <td>0</td>\n", |
||
| 434 | " <td>0</td>\n", |
||
| 435 | " <td>0</td>\n", |
||
| 436 | " <td>0</td>\n", |
||
| 437 | " </tr>\n", |
||
| 438 | " <tr>\n", |
||
| 439 | " <th>ID_000039fa0</th>\n", |
||
| 440 | " <td>0</td>\n", |
||
| 441 | " <td>0</td>\n", |
||
| 442 | " <td>0</td>\n", |
||
| 443 | " <td>0</td>\n", |
||
| 444 | " <td>0</td>\n", |
||
| 445 | " <td>0</td>\n", |
||
| 446 | " </tr>\n", |
||
| 447 | " <tr>\n", |
||
| 448 | " <th>ID_00005679d</th>\n", |
||
| 449 | " <td>0</td>\n", |
||
| 450 | " <td>0</td>\n", |
||
| 451 | " <td>0</td>\n", |
||
| 452 | " <td>0</td>\n", |
||
| 453 | " <td>0</td>\n", |
||
| 454 | " <td>0</td>\n", |
||
| 455 | " </tr>\n", |
||
| 456 | " <tr>\n", |
||
| 457 | " <th>ID_00008ce3c</th>\n", |
||
| 458 | " <td>0</td>\n", |
||
| 459 | " <td>0</td>\n", |
||
| 460 | " <td>0</td>\n", |
||
| 461 | " <td>0</td>\n", |
||
| 462 | " <td>0</td>\n", |
||
| 463 | " <td>0</td>\n", |
||
| 464 | " </tr>\n", |
||
| 465 | " <tr>\n", |
||
| 466 | " <th>ID_0000950d7</th>\n", |
||
| 467 | " <td>0</td>\n", |
||
| 468 | " <td>0</td>\n", |
||
| 469 | " <td>0</td>\n", |
||
| 470 | " <td>0</td>\n", |
||
| 471 | " <td>0</td>\n", |
||
| 472 | " <td>0</td>\n", |
||
| 473 | " </tr>\n", |
||
| 474 | " </tbody>\n", |
||
| 475 | "</table>\n", |
||
| 476 | "</div>" |
||
| 477 | ], |
||
| 478 | "text/plain": [ |
||
| 479 | "label any epidural intraparenchymal intraventricular subarachnoid \\\n", |
||
| 480 | "fid \n", |
||
| 481 | "ID_000012eaf 0 0 0 0 0 \n", |
||
| 482 | "ID_000039fa0 0 0 0 0 0 \n", |
||
| 483 | "ID_00005679d 0 0 0 0 0 \n", |
||
| 484 | "ID_00008ce3c 0 0 0 0 0 \n", |
||
| 485 | "ID_0000950d7 0 0 0 0 0 \n", |
||
| 486 | "\n", |
||
| 487 | "label subdural \n", |
||
| 488 | "fid \n", |
||
| 489 | "ID_000012eaf 0 \n", |
||
| 490 | "ID_000039fa0 0 \n", |
||
| 491 | "ID_00005679d 0 \n", |
||
| 492 | "ID_00008ce3c 0 \n", |
||
| 493 | "ID_0000950d7 0 " |
||
| 494 | ] |
||
| 495 | }, |
||
| 496 | "execution_count": 18, |
||
| 497 | "metadata": {}, |
||
| 498 | "output_type": "execute_result" |
||
| 499 | } |
||
| 500 | ], |
||
| 501 | "source": [ |
||
| 502 | "df_diags.head()" |
||
| 503 | ] |
||
| 504 | }, |
||
| 505 | { |
||
| 506 | "cell_type": "code", |
||
| 507 | "execution_count": 19, |
||
| 508 | "metadata": {}, |
||
| 509 | "outputs": [ |
||
| 510 | { |
||
| 511 | "data": { |
||
| 512 | "text/plain": [ |
||
| 513 | "(752802, 6)" |
||
| 514 | ] |
||
| 515 | }, |
||
| 516 | "execution_count": 19, |
||
| 517 | "metadata": {}, |
||
| 518 | "output_type": "execute_result" |
||
| 519 | } |
||
| 520 | ], |
||
| 521 | "source": [ |
||
| 522 | "df_diags.shape" |
||
| 523 | ] |
||
| 524 | }, |
||
| 525 | { |
||
| 526 | "cell_type": "code", |
||
| 527 | "execution_count": 20, |
||
| 528 | "metadata": {}, |
||
| 529 | "outputs": [], |
||
| 530 | "source": [ |
||
| 531 | "df_diags.reset_index(inplace=True)" |
||
| 532 | ] |
||
| 533 | }, |
||
| 534 | { |
||
| 535 | "cell_type": "code", |
||
| 536 | "execution_count": 21, |
||
| 537 | "metadata": {}, |
||
| 538 | "outputs": [ |
||
| 539 | { |
||
| 540 | "data": { |
||
| 541 | "text/html": [ |
||
| 542 | "<div>\n", |
||
| 543 | "<style scoped>\n", |
||
| 544 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 545 | " vertical-align: middle;\n", |
||
| 546 | " }\n", |
||
| 547 | "\n", |
||
| 548 | " .dataframe tbody tr th {\n", |
||
| 549 | " vertical-align: top;\n", |
||
| 550 | " }\n", |
||
| 551 | "\n", |
||
| 552 | " .dataframe thead th {\n", |
||
| 553 | " text-align: right;\n", |
||
| 554 | " }\n", |
||
| 555 | "</style>\n", |
||
| 556 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 557 | " <thead>\n", |
||
| 558 | " <tr style=\"text-align: right;\">\n", |
||
| 559 | " <th>label</th>\n", |
||
| 560 | " <th>fid</th>\n", |
||
| 561 | " <th>any</th>\n", |
||
| 562 | " <th>epidural</th>\n", |
||
| 563 | " <th>intraparenchymal</th>\n", |
||
| 564 | " <th>intraventricular</th>\n", |
||
| 565 | " <th>subarachnoid</th>\n", |
||
| 566 | " <th>subdural</th>\n", |
||
| 567 | " </tr>\n", |
||
| 568 | " </thead>\n", |
||
| 569 | " <tbody>\n", |
||
| 570 | " <tr>\n", |
||
| 571 | " <th>0</th>\n", |
||
| 572 | " <td>ID_000012eaf</td>\n", |
||
| 573 | " <td>0</td>\n", |
||
| 574 | " <td>0</td>\n", |
||
| 575 | " <td>0</td>\n", |
||
| 576 | " <td>0</td>\n", |
||
| 577 | " <td>0</td>\n", |
||
| 578 | " <td>0</td>\n", |
||
| 579 | " </tr>\n", |
||
| 580 | " <tr>\n", |
||
| 581 | " <th>1</th>\n", |
||
| 582 | " <td>ID_000039fa0</td>\n", |
||
| 583 | " <td>0</td>\n", |
||
| 584 | " <td>0</td>\n", |
||
| 585 | " <td>0</td>\n", |
||
| 586 | " <td>0</td>\n", |
||
| 587 | " <td>0</td>\n", |
||
| 588 | " <td>0</td>\n", |
||
| 589 | " </tr>\n", |
||
| 590 | " <tr>\n", |
||
| 591 | " <th>2</th>\n", |
||
| 592 | " <td>ID_00005679d</td>\n", |
||
| 593 | " <td>0</td>\n", |
||
| 594 | " <td>0</td>\n", |
||
| 595 | " <td>0</td>\n", |
||
| 596 | " <td>0</td>\n", |
||
| 597 | " <td>0</td>\n", |
||
| 598 | " <td>0</td>\n", |
||
| 599 | " </tr>\n", |
||
| 600 | " <tr>\n", |
||
| 601 | " <th>3</th>\n", |
||
| 602 | " <td>ID_00008ce3c</td>\n", |
||
| 603 | " <td>0</td>\n", |
||
| 604 | " <td>0</td>\n", |
||
| 605 | " <td>0</td>\n", |
||
| 606 | " <td>0</td>\n", |
||
| 607 | " <td>0</td>\n", |
||
| 608 | " <td>0</td>\n", |
||
| 609 | " </tr>\n", |
||
| 610 | " <tr>\n", |
||
| 611 | " <th>4</th>\n", |
||
| 612 | " <td>ID_0000950d7</td>\n", |
||
| 613 | " <td>0</td>\n", |
||
| 614 | " <td>0</td>\n", |
||
| 615 | " <td>0</td>\n", |
||
| 616 | " <td>0</td>\n", |
||
| 617 | " <td>0</td>\n", |
||
| 618 | " <td>0</td>\n", |
||
| 619 | " </tr>\n", |
||
| 620 | " </tbody>\n", |
||
| 621 | "</table>\n", |
||
| 622 | "</div>" |
||
| 623 | ], |
||
| 624 | "text/plain": [ |
||
| 625 | "label fid any epidural intraparenchymal intraventricular \\\n", |
||
| 626 | "0 ID_000012eaf 0 0 0 0 \n", |
||
| 627 | "1 ID_000039fa0 0 0 0 0 \n", |
||
| 628 | "2 ID_00005679d 0 0 0 0 \n", |
||
| 629 | "3 ID_00008ce3c 0 0 0 0 \n", |
||
| 630 | "4 ID_0000950d7 0 0 0 0 \n", |
||
| 631 | "\n", |
||
| 632 | "label subarachnoid subdural \n", |
||
| 633 | "0 0 0 \n", |
||
| 634 | "1 0 0 \n", |
||
| 635 | "2 0 0 \n", |
||
| 636 | "3 0 0 \n", |
||
| 637 | "4 0 0 " |
||
| 638 | ] |
||
| 639 | }, |
||
| 640 | "execution_count": 21, |
||
| 641 | "metadata": {}, |
||
| 642 | "output_type": "execute_result" |
||
| 643 | } |
||
| 644 | ], |
||
| 645 | "source": [ |
||
| 646 | "df_diags.head()" |
||
| 647 | ] |
||
| 648 | }, |
||
| 649 | { |
||
| 650 | "cell_type": "code", |
||
| 651 | "execution_count": 22, |
||
| 652 | "metadata": {}, |
||
| 653 | "outputs": [ |
||
| 654 | { |
||
| 655 | "data": { |
||
| 656 | "text/plain": [ |
||
| 657 | "(752802, 7)" |
||
| 658 | ] |
||
| 659 | }, |
||
| 660 | "execution_count": 22, |
||
| 661 | "metadata": {}, |
||
| 662 | "output_type": "execute_result" |
||
| 663 | } |
||
| 664 | ], |
||
| 665 | "source": [ |
||
| 666 | "df_diags.shape" |
||
| 667 | ] |
||
| 668 | }, |
||
| 669 | { |
||
| 670 | "cell_type": "code", |
||
| 671 | "execution_count": 23, |
||
| 672 | "metadata": {}, |
||
| 673 | "outputs": [], |
||
| 674 | "source": [ |
||
| 675 | "df_diags.to_csv(f'data/{stage}_train_diags.csv', index=False)" |
||
| 676 | ] |
||
| 677 | }, |
||
| 678 | { |
||
| 679 | "cell_type": "markdown", |
||
| 680 | "metadata": {}, |
||
| 681 | "source": [ |
||
| 682 | "# Generate fastai-ready csv image file (.png) -> labels\n", |
||
| 683 | "This is needed for early experiments that worked with the .png dataset.\n", |
||
| 684 | "\n", |
||
| 685 | "Generates:\n", |
||
| 686 | "* `train_labels_as_strings.csv`" |
||
| 687 | ] |
||
| 688 | }, |
||
| 689 | { |
||
| 690 | "cell_type": "code", |
||
| 691 | "execution_count": 24, |
||
| 692 | "metadata": {}, |
||
| 693 | "outputs": [], |
||
| 694 | "source": [ |
||
| 695 | "from collections import defaultdict\n", |
||
| 696 | "\n", |
||
| 697 | "d = defaultdict(list)\n", |
||
| 698 | "for fid in df_train.fid.unique(): d[fid]\n", |
||
| 699 | "\n", |
||
| 700 | "for tup in df_train.itertuples():\n", |
||
| 701 | " if tup.probability: d[tup.fid].append(tup.label)" |
||
| 702 | ] |
||
| 703 | }, |
||
| 704 | { |
||
| 705 | "cell_type": "code", |
||
| 706 | "execution_count": 25, |
||
| 707 | "metadata": {}, |
||
| 708 | "outputs": [], |
||
| 709 | "source": [ |
||
| 710 | "ks, vs = [], []\n", |
||
| 711 | "\n", |
||
| 712 | "for k, v in d.items():\n", |
||
| 713 | " ks.append(k), vs.append(' '.join(v))" |
||
| 714 | ] |
||
| 715 | }, |
||
| 716 | { |
||
| 717 | "cell_type": "code", |
||
| 718 | "execution_count": 26, |
||
| 719 | "metadata": {}, |
||
| 720 | "outputs": [], |
||
| 721 | "source": [ |
||
| 722 | "fastai_df = pd.DataFrame(data={'fn': ks, 'labels': vs})" |
||
| 723 | ] |
||
| 724 | }, |
||
| 725 | { |
||
| 726 | "cell_type": "code", |
||
| 727 | "execution_count": 27, |
||
| 728 | "metadata": {}, |
||
| 729 | "outputs": [ |
||
| 730 | { |
||
| 731 | "data": { |
||
| 732 | "text/plain": [ |
||
| 733 | "(752802, 2)" |
||
| 734 | ] |
||
| 735 | }, |
||
| 736 | "execution_count": 27, |
||
| 737 | "metadata": {}, |
||
| 738 | "output_type": "execute_result" |
||
| 739 | } |
||
| 740 | ], |
||
| 741 | "source": [ |
||
| 742 | "fastai_df.shape" |
||
| 743 | ] |
||
| 744 | }, |
||
| 745 | { |
||
| 746 | "cell_type": "code", |
||
| 747 | "execution_count": 28, |
||
| 748 | "metadata": {}, |
||
| 749 | "outputs": [], |
||
| 750 | "source": [ |
||
| 751 | "fastai_df['fn'] += '.png'" |
||
| 752 | ] |
||
| 753 | }, |
||
| 754 | { |
||
| 755 | "cell_type": "code", |
||
| 756 | "execution_count": 29, |
||
| 757 | "metadata": {}, |
||
| 758 | "outputs": [ |
||
| 759 | { |
||
| 760 | "data": { |
||
| 761 | "text/html": [ |
||
| 762 | "<div>\n", |
||
| 763 | "<style scoped>\n", |
||
| 764 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 765 | " vertical-align: middle;\n", |
||
| 766 | " }\n", |
||
| 767 | "\n", |
||
| 768 | " .dataframe tbody tr th {\n", |
||
| 769 | " vertical-align: top;\n", |
||
| 770 | " }\n", |
||
| 771 | "\n", |
||
| 772 | " .dataframe thead th {\n", |
||
| 773 | " text-align: right;\n", |
||
| 774 | " }\n", |
||
| 775 | "</style>\n", |
||
| 776 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 777 | " <thead>\n", |
||
| 778 | " <tr style=\"text-align: right;\">\n", |
||
| 779 | " <th></th>\n", |
||
| 780 | " <th>fn</th>\n", |
||
| 781 | " <th>labels</th>\n", |
||
| 782 | " </tr>\n", |
||
| 783 | " </thead>\n", |
||
| 784 | " <tbody>\n", |
||
| 785 | " <tr>\n", |
||
| 786 | " <th>0</th>\n", |
||
| 787 | " <td>ID_12cadc6af.png</td>\n", |
||
| 788 | " <td></td>\n", |
||
| 789 | " </tr>\n", |
||
| 790 | " <tr>\n", |
||
| 791 | " <th>1</th>\n", |
||
| 792 | " <td>ID_38fd7baa0.png</td>\n", |
||
| 793 | " <td></td>\n", |
||
| 794 | " </tr>\n", |
||
| 795 | " <tr>\n", |
||
| 796 | " <th>2</th>\n", |
||
| 797 | " <td>ID_6c5d82413.png</td>\n", |
||
| 798 | " <td></td>\n", |
||
| 799 | " </tr>\n", |
||
| 800 | " <tr>\n", |
||
| 801 | " <th>3</th>\n", |
||
| 802 | " <td>ID_aec8e68b3.png</td>\n", |
||
| 803 | " <td>subarachnoid any</td>\n", |
||
| 804 | " </tr>\n", |
||
| 805 | " <tr>\n", |
||
| 806 | " <th>4</th>\n", |
||
| 807 | " <td>ID_4d9209c7c.png</td>\n", |
||
| 808 | " <td></td>\n", |
||
| 809 | " </tr>\n", |
||
| 810 | " </tbody>\n", |
||
| 811 | "</table>\n", |
||
| 812 | "</div>" |
||
| 813 | ], |
||
| 814 | "text/plain": [ |
||
| 815 | " fn labels\n", |
||
| 816 | "0 ID_12cadc6af.png \n", |
||
| 817 | "1 ID_38fd7baa0.png \n", |
||
| 818 | "2 ID_6c5d82413.png \n", |
||
| 819 | "3 ID_aec8e68b3.png subarachnoid any\n", |
||
| 820 | "4 ID_4d9209c7c.png " |
||
| 821 | ] |
||
| 822 | }, |
||
| 823 | "execution_count": 29, |
||
| 824 | "metadata": {}, |
||
| 825 | "output_type": "execute_result" |
||
| 826 | } |
||
| 827 | ], |
||
| 828 | "source": [ |
||
| 829 | "fastai_df.head()" |
||
| 830 | ] |
||
| 831 | }, |
||
| 832 | { |
||
| 833 | "cell_type": "code", |
||
| 834 | "execution_count": 30, |
||
| 835 | "metadata": {}, |
||
| 836 | "outputs": [], |
||
| 837 | "source": [ |
||
| 838 | "fastai_df.to_csv(f'data/{stage}_train_labels_as_strings.csv', index=False)" |
||
| 839 | ] |
||
| 840 | }, |
||
| 841 | { |
||
| 842 | "cell_type": "markdown", |
||
| 843 | "metadata": {}, |
||
| 844 | "source": [ |
||
| 845 | "# Tabulate dicom data\n", |
||
| 846 | "Generates:\n", |
||
| 847 | "* `train_dicom.csv`\n", |
||
| 848 | "* `test_dicom.csv`" |
||
| 849 | ] |
||
| 850 | }, |
||
| 851 | { |
||
| 852 | "cell_type": "code", |
||
| 853 | "execution_count": 31, |
||
| 854 | "metadata": {}, |
||
| 855 | "outputs": [], |
||
| 856 | "source": [ |
||
| 857 | "cols_i_want = ['BitsAllocated', 'BitsStored', 'Columns', 'HighBit', 'ImageOrientationPatient', \n", |
||
| 858 | " 'ImagePositionPatient', 'Modality', 'PatientID', 'PhotometricInterpretation', \n", |
||
| 859 | " 'PixelRepresentation', 'PixelSpacing', 'RescaleIntercept', 'RescaleSlope', \n", |
||
| 860 | " 'Rows', 'SOPInstanceUID', 'SamplesPerPixel', 'SeriesInstanceUID', 'StudyID', \n", |
||
| 861 | " 'StudyInstanceUID', 'WindowCenter', 'WindowWidth']\n", |
||
| 862 | "useless_cols = [ 'PixelData' ]" |
||
| 863 | ] |
||
| 864 | }, |
||
| 865 | { |
||
| 866 | "cell_type": "code", |
||
| 867 | "execution_count": 32, |
||
| 868 | "metadata": {}, |
||
| 869 | "outputs": [], |
||
| 870 | "source": [ |
||
| 871 | "train_dcm_list = list(Path(f'data/unzip/{stage}_train_images').glob('*.dcm'))" |
||
| 872 | ] |
||
| 873 | }, |
||
| 874 | { |
||
| 875 | "cell_type": "code", |
||
| 876 | "execution_count": 33, |
||
| 877 | "metadata": {}, |
||
| 878 | "outputs": [], |
||
| 879 | "source": [ |
||
| 880 | "f = train_dcm_list[0]" |
||
| 881 | ] |
||
| 882 | }, |
||
| 883 | { |
||
| 884 | "cell_type": "code", |
||
| 885 | "execution_count": 34, |
||
| 886 | "metadata": {}, |
||
| 887 | "outputs": [ |
||
| 888 | { |
||
| 889 | "data": { |
||
| 890 | "text/plain": [ |
||
| 891 | "(0008, 0018) SOP Instance UID UI: ID_000012eaf\n", |
||
| 892 | "(0008, 0060) Modality CS: 'CT'\n", |
||
| 893 | "(0010, 0020) Patient ID LO: 'ID_f15c0eee'\n", |
||
| 894 | "(0020, 000d) Study Instance UID UI: ID_30ea2b02d4\n", |
||
| 895 | "(0020, 000e) Series Instance UID UI: ID_0ab5820b2a\n", |
||
| 896 | "(0020, 0010) Study ID SH: ''\n", |
||
| 897 | "(0020, 0032) Image Position (Patient) DS: ['-125.000000', '-115.897980', '77.970825']\n", |
||
| 898 | "(0020, 0037) Image Orientation (Patient) DS: ['1.000000', '0.000000', '0.000000', '0.000000', '0.927184', '-0.374607']\n", |
||
| 899 | "(0028, 0002) Samples per Pixel US: 1\n", |
||
| 900 | "(0028, 0004) Photometric Interpretation CS: 'MONOCHROME2'\n", |
||
| 901 | "(0028, 0010) Rows US: 512\n", |
||
| 902 | "(0028, 0011) Columns US: 512\n", |
||
| 903 | "(0028, 0030) Pixel Spacing DS: ['0.488281', '0.488281']\n", |
||
| 904 | "(0028, 0100) Bits Allocated US: 16\n", |
||
| 905 | "(0028, 0101) Bits Stored US: 16\n", |
||
| 906 | "(0028, 0102) High Bit US: 15\n", |
||
| 907 | "(0028, 0103) Pixel Representation US: 1\n", |
||
| 908 | "(0028, 1050) Window Center DS: \"30\"\n", |
||
| 909 | "(0028, 1051) Window Width DS: \"80\"\n", |
||
| 910 | "(0028, 1052) Rescale Intercept DS: \"-1024\"\n", |
||
| 911 | "(0028, 1053) Rescale Slope DS: \"1\"\n", |
||
| 912 | "(7fe0, 0010) Pixel Data OW: Array of 524288 elements" |
||
| 913 | ] |
||
| 914 | }, |
||
| 915 | "execution_count": 34, |
||
| 916 | "metadata": {}, |
||
| 917 | "output_type": "execute_result" |
||
| 918 | } |
||
| 919 | ], |
||
| 920 | "source": [ |
||
| 921 | "dicom = pydicom.dcmread(str(f))\n", |
||
| 922 | "dicom" |
||
| 923 | ] |
||
| 924 | }, |
||
| 925 | { |
||
| 926 | "cell_type": "code", |
||
| 927 | "execution_count": 35, |
||
| 928 | "metadata": {}, |
||
| 929 | "outputs": [ |
||
| 930 | { |
||
| 931 | "data": { |
||
| 932 | "text/plain": [ |
||
| 933 | "'ID_f15c0eee'" |
||
| 934 | ] |
||
| 935 | }, |
||
| 936 | "execution_count": 35, |
||
| 937 | "metadata": {}, |
||
| 938 | "output_type": "execute_result" |
||
| 939 | } |
||
| 940 | ], |
||
| 941 | "source": [ |
||
| 942 | "dicom.data_element('PatientID').value" |
||
| 943 | ] |
||
| 944 | }, |
||
| 945 | { |
||
| 946 | "cell_type": "code", |
||
| 947 | "execution_count": 36, |
||
| 948 | "metadata": {}, |
||
| 949 | "outputs": [ |
||
| 950 | { |
||
| 951 | "data": { |
||
| 952 | "text/plain": [ |
||
| 953 | "['-125.000000', '-115.897980', '77.970825']" |
||
| 954 | ] |
||
| 955 | }, |
||
| 956 | "execution_count": 36, |
||
| 957 | "metadata": {}, |
||
| 958 | "output_type": "execute_result" |
||
| 959 | } |
||
| 960 | ], |
||
| 961 | "source": [ |
||
| 962 | "ipp = dicom.data_element('ImagePositionPatient').value\n", |
||
| 963 | "ipp" |
||
| 964 | ] |
||
| 965 | }, |
||
| 966 | { |
||
| 967 | "cell_type": "code", |
||
| 968 | "execution_count": 37, |
||
| 969 | "metadata": {}, |
||
| 970 | "outputs": [ |
||
| 971 | { |
||
| 972 | "data": { |
||
| 973 | "text/plain": [ |
||
| 974 | "\"-125.000000\"" |
||
| 975 | ] |
||
| 976 | }, |
||
| 977 | "execution_count": 37, |
||
| 978 | "metadata": {}, |
||
| 979 | "output_type": "execute_result" |
||
| 980 | } |
||
| 981 | ], |
||
| 982 | "source": [ |
||
| 983 | "ipp[0]" |
||
| 984 | ] |
||
| 985 | }, |
||
| 986 | { |
||
| 987 | "cell_type": "code", |
||
| 988 | "execution_count": 38, |
||
| 989 | "metadata": {}, |
||
| 990 | "outputs": [ |
||
| 991 | { |
||
| 992 | "data": { |
||
| 993 | "text/plain": [ |
||
| 994 | "pydicom.multival.MultiValue" |
||
| 995 | ] |
||
| 996 | }, |
||
| 997 | "execution_count": 38, |
||
| 998 | "metadata": {}, |
||
| 999 | "output_type": "execute_result" |
||
| 1000 | } |
||
| 1001 | ], |
||
| 1002 | "source": [ |
||
| 1003 | "type(ipp)" |
||
| 1004 | ] |
||
| 1005 | }, |
||
| 1006 | { |
||
| 1007 | "cell_type": "code", |
||
| 1008 | "execution_count": 39, |
||
| 1009 | "metadata": {}, |
||
| 1010 | "outputs": [], |
||
| 1011 | "source": [ |
||
| 1012 | "def tabulate_dicom_data(file_list):\n", |
||
| 1013 | " l = []\n", |
||
| 1014 | " for f in file_list:\n", |
||
| 1015 | " dicom = pydicom.dcmread(str(f))\n", |
||
| 1016 | " d = {}\n", |
||
| 1017 | " for s in cols_i_want:\n", |
||
| 1018 | " v = dicom.data_element(s).value\n", |
||
| 1019 | " if isinstance(v, pydicom.multival.MultiValue):\n", |
||
| 1020 | " for i in range(len(v)):\n", |
||
| 1021 | " d[s + '_' + str(i)] = v[i]\n", |
||
| 1022 | " else:\n", |
||
| 1023 | " d[s] = v\n", |
||
| 1024 | " l.append(d)\n", |
||
| 1025 | " \n", |
||
| 1026 | " return l\n" |
||
| 1027 | ] |
||
| 1028 | }, |
||
| 1029 | { |
||
| 1030 | "cell_type": "code", |
||
| 1031 | "execution_count": 40, |
||
| 1032 | "metadata": {}, |
||
| 1033 | "outputs": [], |
||
| 1034 | "source": [ |
||
| 1035 | "with ProcessPoolExecutor(max_workers=32) as e:\n", |
||
| 1036 | " l = list(itertools.chain.from_iterable(e.map(tabulate_dicom_data, np.array_split(train_dcm_list, 32))))" |
||
| 1037 | ] |
||
| 1038 | }, |
||
| 1039 | { |
||
| 1040 | "cell_type": "code", |
||
| 1041 | "execution_count": 41, |
||
| 1042 | "metadata": {}, |
||
| 1043 | "outputs": [], |
||
| 1044 | "source": [ |
||
| 1045 | "df_train_dicom = pd.DataFrame(l)" |
||
| 1046 | ] |
||
| 1047 | }, |
||
| 1048 | { |
||
| 1049 | "cell_type": "code", |
||
| 1050 | "execution_count": 42, |
||
| 1051 | "metadata": {}, |
||
| 1052 | "outputs": [ |
||
| 1053 | { |
||
| 1054 | "data": { |
||
| 1055 | "text/html": [ |
||
| 1056 | "<div>\n", |
||
| 1057 | "<style scoped>\n", |
||
| 1058 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 1059 | " vertical-align: middle;\n", |
||
| 1060 | " }\n", |
||
| 1061 | "\n", |
||
| 1062 | " .dataframe tbody tr th {\n", |
||
| 1063 | " vertical-align: top;\n", |
||
| 1064 | " }\n", |
||
| 1065 | "\n", |
||
| 1066 | " .dataframe thead th {\n", |
||
| 1067 | " text-align: right;\n", |
||
| 1068 | " }\n", |
||
| 1069 | "</style>\n", |
||
| 1070 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 1071 | " <thead>\n", |
||
| 1072 | " <tr style=\"text-align: right;\">\n", |
||
| 1073 | " <th></th>\n", |
||
| 1074 | " <th>BitsAllocated</th>\n", |
||
| 1075 | " <th>BitsStored</th>\n", |
||
| 1076 | " <th>Columns</th>\n", |
||
| 1077 | " <th>HighBit</th>\n", |
||
| 1078 | " <th>ImageOrientationPatient_0</th>\n", |
||
| 1079 | " <th>ImageOrientationPatient_1</th>\n", |
||
| 1080 | " <th>ImageOrientationPatient_2</th>\n", |
||
| 1081 | " <th>ImageOrientationPatient_3</th>\n", |
||
| 1082 | " <th>ImageOrientationPatient_4</th>\n", |
||
| 1083 | " <th>ImageOrientationPatient_5</th>\n", |
||
| 1084 | " <th>...</th>\n", |
||
| 1085 | " <th>SamplesPerPixel</th>\n", |
||
| 1086 | " <th>SeriesInstanceUID</th>\n", |
||
| 1087 | " <th>StudyID</th>\n", |
||
| 1088 | " <th>StudyInstanceUID</th>\n", |
||
| 1089 | " <th>WindowCenter</th>\n", |
||
| 1090 | " <th>WindowCenter_0</th>\n", |
||
| 1091 | " <th>WindowCenter_1</th>\n", |
||
| 1092 | " <th>WindowWidth</th>\n", |
||
| 1093 | " <th>WindowWidth_0</th>\n", |
||
| 1094 | " <th>WindowWidth_1</th>\n", |
||
| 1095 | " </tr>\n", |
||
| 1096 | " </thead>\n", |
||
| 1097 | " <tbody>\n", |
||
| 1098 | " <tr>\n", |
||
| 1099 | " <th>0</th>\n", |
||
| 1100 | " <td>16</td>\n", |
||
| 1101 | " <td>16</td>\n", |
||
| 1102 | " <td>512</td>\n", |
||
| 1103 | " <td>15</td>\n", |
||
| 1104 | " <td>1.0</td>\n", |
||
| 1105 | " <td>0.0</td>\n", |
||
| 1106 | " <td>0.0</td>\n", |
||
| 1107 | " <td>0.0</td>\n", |
||
| 1108 | " <td>0.927184</td>\n", |
||
| 1109 | " <td>-0.374607</td>\n", |
||
| 1110 | " <td>...</td>\n", |
||
| 1111 | " <td>1</td>\n", |
||
| 1112 | " <td>ID_0ab5820b2a</td>\n", |
||
| 1113 | " <td></td>\n", |
||
| 1114 | " <td>ID_30ea2b02d4</td>\n", |
||
| 1115 | " <td>30.0</td>\n", |
||
| 1116 | " <td>NaN</td>\n", |
||
| 1117 | " <td>NaN</td>\n", |
||
| 1118 | " <td>80.0</td>\n", |
||
| 1119 | " <td>NaN</td>\n", |
||
| 1120 | " <td>NaN</td>\n", |
||
| 1121 | " </tr>\n", |
||
| 1122 | " <tr>\n", |
||
| 1123 | " <th>1</th>\n", |
||
| 1124 | " <td>16</td>\n", |
||
| 1125 | " <td>16</td>\n", |
||
| 1126 | " <td>512</td>\n", |
||
| 1127 | " <td>15</td>\n", |
||
| 1128 | " <td>1.0</td>\n", |
||
| 1129 | " <td>0.0</td>\n", |
||
| 1130 | " <td>0.0</td>\n", |
||
| 1131 | " <td>0.0</td>\n", |
||
| 1132 | " <td>0.968148</td>\n", |
||
| 1133 | " <td>-0.250380</td>\n", |
||
| 1134 | " <td>...</td>\n", |
||
| 1135 | " <td>1</td>\n", |
||
| 1136 | " <td>ID_5f8484c3e0</td>\n", |
||
| 1137 | " <td></td>\n", |
||
| 1138 | " <td>ID_134d398b61</td>\n", |
||
| 1139 | " <td>30.0</td>\n", |
||
| 1140 | " <td>NaN</td>\n", |
||
| 1141 | " <td>NaN</td>\n", |
||
| 1142 | " <td>80.0</td>\n", |
||
| 1143 | " <td>NaN</td>\n", |
||
| 1144 | " <td>NaN</td>\n", |
||
| 1145 | " </tr>\n", |
||
| 1146 | " <tr>\n", |
||
| 1147 | " <th>2</th>\n", |
||
| 1148 | " <td>16</td>\n", |
||
| 1149 | " <td>16</td>\n", |
||
| 1150 | " <td>512</td>\n", |
||
| 1151 | " <td>15</td>\n", |
||
| 1152 | " <td>1.0</td>\n", |
||
| 1153 | " <td>0.0</td>\n", |
||
| 1154 | " <td>0.0</td>\n", |
||
| 1155 | " <td>0.0</td>\n", |
||
| 1156 | " <td>1.000000</td>\n", |
||
| 1157 | " <td>0.000000</td>\n", |
||
| 1158 | " <td>...</td>\n", |
||
| 1159 | " <td>1</td>\n", |
||
| 1160 | " <td>ID_203cd6ec46</td>\n", |
||
| 1161 | " <td></td>\n", |
||
| 1162 | " <td>ID_b5c26cda09</td>\n", |
||
| 1163 | " <td>50.0</td>\n", |
||
| 1164 | " <td>NaN</td>\n", |
||
| 1165 | " <td>NaN</td>\n", |
||
| 1166 | " <td>100.0</td>\n", |
||
| 1167 | " <td>NaN</td>\n", |
||
| 1168 | " <td>NaN</td>\n", |
||
| 1169 | " </tr>\n", |
||
| 1170 | " <tr>\n", |
||
| 1171 | " <th>3</th>\n", |
||
| 1172 | " <td>16</td>\n", |
||
| 1173 | " <td>12</td>\n", |
||
| 1174 | " <td>512</td>\n", |
||
| 1175 | " <td>11</td>\n", |
||
| 1176 | " <td>1.0</td>\n", |
||
| 1177 | " <td>0.0</td>\n", |
||
| 1178 | " <td>0.0</td>\n", |
||
| 1179 | " <td>0.0</td>\n", |
||
| 1180 | " <td>0.994522</td>\n", |
||
| 1181 | " <td>0.104528</td>\n", |
||
| 1182 | " <td>...</td>\n", |
||
| 1183 | " <td>1</td>\n", |
||
| 1184 | " <td>ID_3780d48b28</td>\n", |
||
| 1185 | " <td></td>\n", |
||
| 1186 | " <td>ID_974735bf79</td>\n", |
||
| 1187 | " <td>NaN</td>\n", |
||
| 1188 | " <td>40.0</td>\n", |
||
| 1189 | " <td>40.0</td>\n", |
||
| 1190 | " <td>NaN</td>\n", |
||
| 1191 | " <td>80.0</td>\n", |
||
| 1192 | " <td>80.0</td>\n", |
||
| 1193 | " </tr>\n", |
||
| 1194 | " <tr>\n", |
||
| 1195 | " <th>4</th>\n", |
||
| 1196 | " <td>16</td>\n", |
||
| 1197 | " <td>16</td>\n", |
||
| 1198 | " <td>512</td>\n", |
||
| 1199 | " <td>15</td>\n", |
||
| 1200 | " <td>1.0</td>\n", |
||
| 1201 | " <td>0.0</td>\n", |
||
| 1202 | " <td>0.0</td>\n", |
||
| 1203 | " <td>0.0</td>\n", |
||
| 1204 | " <td>1.000000</td>\n", |
||
| 1205 | " <td>0.000000</td>\n", |
||
| 1206 | " <td>...</td>\n", |
||
| 1207 | " <td>1</td>\n", |
||
| 1208 | " <td>ID_84296c3845</td>\n", |
||
| 1209 | " <td></td>\n", |
||
| 1210 | " <td>ID_8881b1c4b1</td>\n", |
||
| 1211 | " <td>35.0</td>\n", |
||
| 1212 | " <td>NaN</td>\n", |
||
| 1213 | " <td>NaN</td>\n", |
||
| 1214 | " <td>135.0</td>\n", |
||
| 1215 | " <td>NaN</td>\n", |
||
| 1216 | " <td>NaN</td>\n", |
||
| 1217 | " </tr>\n", |
||
| 1218 | " </tbody>\n", |
||
| 1219 | "</table>\n", |
||
| 1220 | "<p>5 rows × 33 columns</p>\n", |
||
| 1221 | "</div>" |
||
| 1222 | ], |
||
| 1223 | "text/plain": [ |
||
| 1224 | " BitsAllocated BitsStored Columns HighBit ImageOrientationPatient_0 \\\n", |
||
| 1225 | "0 16 16 512 15 1.0 \n", |
||
| 1226 | "1 16 16 512 15 1.0 \n", |
||
| 1227 | "2 16 16 512 15 1.0 \n", |
||
| 1228 | "3 16 12 512 11 1.0 \n", |
||
| 1229 | "4 16 16 512 15 1.0 \n", |
||
| 1230 | "\n", |
||
| 1231 | " ImageOrientationPatient_1 ImageOrientationPatient_2 \\\n", |
||
| 1232 | "0 0.0 0.0 \n", |
||
| 1233 | "1 0.0 0.0 \n", |
||
| 1234 | "2 0.0 0.0 \n", |
||
| 1235 | "3 0.0 0.0 \n", |
||
| 1236 | "4 0.0 0.0 \n", |
||
| 1237 | "\n", |
||
| 1238 | " ImageOrientationPatient_3 ImageOrientationPatient_4 \\\n", |
||
| 1239 | "0 0.0 0.927184 \n", |
||
| 1240 | "1 0.0 0.968148 \n", |
||
| 1241 | "2 0.0 1.000000 \n", |
||
| 1242 | "3 0.0 0.994522 \n", |
||
| 1243 | "4 0.0 1.000000 \n", |
||
| 1244 | "\n", |
||
| 1245 | " ImageOrientationPatient_5 ... SamplesPerPixel SeriesInstanceUID \\\n", |
||
| 1246 | "0 -0.374607 ... 1 ID_0ab5820b2a \n", |
||
| 1247 | "1 -0.250380 ... 1 ID_5f8484c3e0 \n", |
||
| 1248 | "2 0.000000 ... 1 ID_203cd6ec46 \n", |
||
| 1249 | "3 0.104528 ... 1 ID_3780d48b28 \n", |
||
| 1250 | "4 0.000000 ... 1 ID_84296c3845 \n", |
||
| 1251 | "\n", |
||
| 1252 | " StudyID StudyInstanceUID WindowCenter WindowCenter_0 WindowCenter_1 \\\n", |
||
| 1253 | "0 ID_30ea2b02d4 30.0 NaN NaN \n", |
||
| 1254 | "1 ID_134d398b61 30.0 NaN NaN \n", |
||
| 1255 | "2 ID_b5c26cda09 50.0 NaN NaN \n", |
||
| 1256 | "3 ID_974735bf79 NaN 40.0 40.0 \n", |
||
| 1257 | "4 ID_8881b1c4b1 35.0 NaN NaN \n", |
||
| 1258 | "\n", |
||
| 1259 | " WindowWidth WindowWidth_0 WindowWidth_1 \n", |
||
| 1260 | "0 80.0 NaN NaN \n", |
||
| 1261 | "1 80.0 NaN NaN \n", |
||
| 1262 | "2 100.0 NaN NaN \n", |
||
| 1263 | "3 NaN 80.0 80.0 \n", |
||
| 1264 | "4 135.0 NaN NaN \n", |
||
| 1265 | "\n", |
||
| 1266 | "[5 rows x 33 columns]" |
||
| 1267 | ] |
||
| 1268 | }, |
||
| 1269 | "execution_count": 42, |
||
| 1270 | "metadata": {}, |
||
| 1271 | "output_type": "execute_result" |
||
| 1272 | } |
||
| 1273 | ], |
||
| 1274 | "source": [ |
||
| 1275 | "df_train_dicom.head()" |
||
| 1276 | ] |
||
| 1277 | }, |
||
| 1278 | { |
||
| 1279 | "cell_type": "code", |
||
| 1280 | "execution_count": 43, |
||
| 1281 | "metadata": {}, |
||
| 1282 | "outputs": [], |
||
| 1283 | "source": [ |
||
| 1284 | "df_train_dicom.to_csv(f'data/{stage}_train_dicom.csv')" |
||
| 1285 | ] |
||
| 1286 | }, |
||
| 1287 | { |
||
| 1288 | "cell_type": "code", |
||
| 1289 | "execution_count": 44, |
||
| 1290 | "metadata": {}, |
||
| 1291 | "outputs": [], |
||
| 1292 | "source": [ |
||
| 1293 | "test_dcm_list = list(Path(f'data/unzip/{stage}_test_images').glob('*.dcm'))\n", |
||
| 1294 | "with ProcessPoolExecutor(max_workers=32) as e:\n", |
||
| 1295 | " l = list(itertools.chain.from_iterable(e.map(tabulate_dicom_data, np.array_split(test_dcm_list, 32))))" |
||
| 1296 | ] |
||
| 1297 | }, |
||
| 1298 | { |
||
| 1299 | "cell_type": "code", |
||
| 1300 | "execution_count": 45, |
||
| 1301 | "metadata": {}, |
||
| 1302 | "outputs": [ |
||
| 1303 | { |
||
| 1304 | "data": { |
||
| 1305 | "text/html": [ |
||
| 1306 | "<div>\n", |
||
| 1307 | "<style scoped>\n", |
||
| 1308 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 1309 | " vertical-align: middle;\n", |
||
| 1310 | " }\n", |
||
| 1311 | "\n", |
||
| 1312 | " .dataframe tbody tr th {\n", |
||
| 1313 | " vertical-align: top;\n", |
||
| 1314 | " }\n", |
||
| 1315 | "\n", |
||
| 1316 | " .dataframe thead th {\n", |
||
| 1317 | " text-align: right;\n", |
||
| 1318 | " }\n", |
||
| 1319 | "</style>\n", |
||
| 1320 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 1321 | " <thead>\n", |
||
| 1322 | " <tr style=\"text-align: right;\">\n", |
||
| 1323 | " <th></th>\n", |
||
| 1324 | " <th>BitsAllocated</th>\n", |
||
| 1325 | " <th>BitsStored</th>\n", |
||
| 1326 | " <th>Columns</th>\n", |
||
| 1327 | " <th>HighBit</th>\n", |
||
| 1328 | " <th>ImageOrientationPatient_0</th>\n", |
||
| 1329 | " <th>ImageOrientationPatient_1</th>\n", |
||
| 1330 | " <th>ImageOrientationPatient_2</th>\n", |
||
| 1331 | " <th>ImageOrientationPatient_3</th>\n", |
||
| 1332 | " <th>ImageOrientationPatient_4</th>\n", |
||
| 1333 | " <th>ImageOrientationPatient_5</th>\n", |
||
| 1334 | " <th>...</th>\n", |
||
| 1335 | " <th>SamplesPerPixel</th>\n", |
||
| 1336 | " <th>SeriesInstanceUID</th>\n", |
||
| 1337 | " <th>StudyID</th>\n", |
||
| 1338 | " <th>StudyInstanceUID</th>\n", |
||
| 1339 | " <th>WindowCenter</th>\n", |
||
| 1340 | " <th>WindowCenter_0</th>\n", |
||
| 1341 | " <th>WindowCenter_1</th>\n", |
||
| 1342 | " <th>WindowWidth</th>\n", |
||
| 1343 | " <th>WindowWidth_0</th>\n", |
||
| 1344 | " <th>WindowWidth_1</th>\n", |
||
| 1345 | " </tr>\n", |
||
| 1346 | " </thead>\n", |
||
| 1347 | " <tbody>\n", |
||
| 1348 | " <tr>\n", |
||
| 1349 | " <th>0</th>\n", |
||
| 1350 | " <td>16</td>\n", |
||
| 1351 | " <td>12</td>\n", |
||
| 1352 | " <td>512</td>\n", |
||
| 1353 | " <td>11</td>\n", |
||
| 1354 | " <td>1.0</td>\n", |
||
| 1355 | " <td>0.0</td>\n", |
||
| 1356 | " <td>0.0</td>\n", |
||
| 1357 | " <td>0.0</td>\n", |
||
| 1358 | " <td>0.981627</td>\n", |
||
| 1359 | " <td>-0.190809</td>\n", |
||
| 1360 | " <td>...</td>\n", |
||
| 1361 | " <td>1</td>\n", |
||
| 1362 | " <td>ID_4d28912ba6</td>\n", |
||
| 1363 | " <td></td>\n", |
||
| 1364 | " <td>ID_1f6d1e8aeb</td>\n", |
||
| 1365 | " <td>NaN</td>\n", |
||
| 1366 | " <td>40.0</td>\n", |
||
| 1367 | " <td>40.0</td>\n", |
||
| 1368 | " <td>NaN</td>\n", |
||
| 1369 | " <td>80.0</td>\n", |
||
| 1370 | " <td>80.0</td>\n", |
||
| 1371 | " </tr>\n", |
||
| 1372 | " <tr>\n", |
||
| 1373 | " <th>1</th>\n", |
||
| 1374 | " <td>16</td>\n", |
||
| 1375 | " <td>16</td>\n", |
||
| 1376 | " <td>512</td>\n", |
||
| 1377 | " <td>15</td>\n", |
||
| 1378 | " <td>1.0</td>\n", |
||
| 1379 | " <td>0.0</td>\n", |
||
| 1380 | " <td>0.0</td>\n", |
||
| 1381 | " <td>0.0</td>\n", |
||
| 1382 | " <td>0.987688</td>\n", |
||
| 1383 | " <td>-0.156434</td>\n", |
||
| 1384 | " <td>...</td>\n", |
||
| 1385 | " <td>1</td>\n", |
||
| 1386 | " <td>ID_acabdeee86</td>\n", |
||
| 1387 | " <td></td>\n", |
||
| 1388 | " <td>ID_4a8d7ec19f</td>\n", |
||
| 1389 | " <td>30.0</td>\n", |
||
| 1390 | " <td>NaN</td>\n", |
||
| 1391 | " <td>NaN</td>\n", |
||
| 1392 | " <td>80.0</td>\n", |
||
| 1393 | " <td>NaN</td>\n", |
||
| 1394 | " <td>NaN</td>\n", |
||
| 1395 | " </tr>\n", |
||
| 1396 | " <tr>\n", |
||
| 1397 | " <th>2</th>\n", |
||
| 1398 | " <td>16</td>\n", |
||
| 1399 | " <td>16</td>\n", |
||
| 1400 | " <td>512</td>\n", |
||
| 1401 | " <td>15</td>\n", |
||
| 1402 | " <td>1.0</td>\n", |
||
| 1403 | " <td>0.0</td>\n", |
||
| 1404 | " <td>0.0</td>\n", |
||
| 1405 | " <td>0.0</td>\n", |
||
| 1406 | " <td>0.927184</td>\n", |
||
| 1407 | " <td>-0.374607</td>\n", |
||
| 1408 | " <td>...</td>\n", |
||
| 1409 | " <td>1</td>\n", |
||
| 1410 | " <td>ID_d00cee7f0c</td>\n", |
||
| 1411 | " <td></td>\n", |
||
| 1412 | " <td>ID_a6ca244172</td>\n", |
||
| 1413 | " <td>30.0</td>\n", |
||
| 1414 | " <td>NaN</td>\n", |
||
| 1415 | " <td>NaN</td>\n", |
||
| 1416 | " <td>80.0</td>\n", |
||
| 1417 | " <td>NaN</td>\n", |
||
| 1418 | " <td>NaN</td>\n", |
||
| 1419 | " </tr>\n", |
||
| 1420 | " <tr>\n", |
||
| 1421 | " <th>3</th>\n", |
||
| 1422 | " <td>16</td>\n", |
||
| 1423 | " <td>16</td>\n", |
||
| 1424 | " <td>512</td>\n", |
||
| 1425 | " <td>15</td>\n", |
||
| 1426 | " <td>1.0</td>\n", |
||
| 1427 | " <td>0.0</td>\n", |
||
| 1428 | " <td>0.0</td>\n", |
||
| 1429 | " <td>0.0</td>\n", |
||
| 1430 | " <td>0.986286</td>\n", |
||
| 1431 | " <td>-0.165048</td>\n", |
||
| 1432 | " <td>...</td>\n", |
||
| 1433 | " <td>1</td>\n", |
||
| 1434 | " <td>ID_a52a0112d5</td>\n", |
||
| 1435 | " <td></td>\n", |
||
| 1436 | " <td>ID_fa950a03af</td>\n", |
||
| 1437 | " <td>30.0</td>\n", |
||
| 1438 | " <td>NaN</td>\n", |
||
| 1439 | " <td>NaN</td>\n", |
||
| 1440 | " <td>80.0</td>\n", |
||
| 1441 | " <td>NaN</td>\n", |
||
| 1442 | " <td>NaN</td>\n", |
||
| 1443 | " </tr>\n", |
||
| 1444 | " <tr>\n", |
||
| 1445 | " <th>4</th>\n", |
||
| 1446 | " <td>16</td>\n", |
||
| 1447 | " <td>12</td>\n", |
||
| 1448 | " <td>512</td>\n", |
||
| 1449 | " <td>11</td>\n", |
||
| 1450 | " <td>1.0</td>\n", |
||
| 1451 | " <td>0.0</td>\n", |
||
| 1452 | " <td>0.0</td>\n", |
||
| 1453 | " <td>0.0</td>\n", |
||
| 1454 | " <td>1.000000</td>\n", |
||
| 1455 | " <td>0.000000</td>\n", |
||
| 1456 | " <td>...</td>\n", |
||
| 1457 | " <td>1</td>\n", |
||
| 1458 | " <td>ID_f552d3b922</td>\n", |
||
| 1459 | " <td></td>\n", |
||
| 1460 | " <td>ID_965d8b3d8e</td>\n", |
||
| 1461 | " <td>NaN</td>\n", |
||
| 1462 | " <td>36.0</td>\n", |
||
| 1463 | " <td>36.0</td>\n", |
||
| 1464 | " <td>NaN</td>\n", |
||
| 1465 | " <td>80.0</td>\n", |
||
| 1466 | " <td>80.0</td>\n", |
||
| 1467 | " </tr>\n", |
||
| 1468 | " </tbody>\n", |
||
| 1469 | "</table>\n", |
||
| 1470 | "<p>5 rows × 33 columns</p>\n", |
||
| 1471 | "</div>" |
||
| 1472 | ], |
||
| 1473 | "text/plain": [ |
||
| 1474 | " BitsAllocated BitsStored Columns HighBit ImageOrientationPatient_0 \\\n", |
||
| 1475 | "0 16 12 512 11 1.0 \n", |
||
| 1476 | "1 16 16 512 15 1.0 \n", |
||
| 1477 | "2 16 16 512 15 1.0 \n", |
||
| 1478 | "3 16 16 512 15 1.0 \n", |
||
| 1479 | "4 16 12 512 11 1.0 \n", |
||
| 1480 | "\n", |
||
| 1481 | " ImageOrientationPatient_1 ImageOrientationPatient_2 \\\n", |
||
| 1482 | "0 0.0 0.0 \n", |
||
| 1483 | "1 0.0 0.0 \n", |
||
| 1484 | "2 0.0 0.0 \n", |
||
| 1485 | "3 0.0 0.0 \n", |
||
| 1486 | "4 0.0 0.0 \n", |
||
| 1487 | "\n", |
||
| 1488 | " ImageOrientationPatient_3 ImageOrientationPatient_4 \\\n", |
||
| 1489 | "0 0.0 0.981627 \n", |
||
| 1490 | "1 0.0 0.987688 \n", |
||
| 1491 | "2 0.0 0.927184 \n", |
||
| 1492 | "3 0.0 0.986286 \n", |
||
| 1493 | "4 0.0 1.000000 \n", |
||
| 1494 | "\n", |
||
| 1495 | " ImageOrientationPatient_5 ... SamplesPerPixel SeriesInstanceUID \\\n", |
||
| 1496 | "0 -0.190809 ... 1 ID_4d28912ba6 \n", |
||
| 1497 | "1 -0.156434 ... 1 ID_acabdeee86 \n", |
||
| 1498 | "2 -0.374607 ... 1 ID_d00cee7f0c \n", |
||
| 1499 | "3 -0.165048 ... 1 ID_a52a0112d5 \n", |
||
| 1500 | "4 0.000000 ... 1 ID_f552d3b922 \n", |
||
| 1501 | "\n", |
||
| 1502 | " StudyID StudyInstanceUID WindowCenter WindowCenter_0 WindowCenter_1 \\\n", |
||
| 1503 | "0 ID_1f6d1e8aeb NaN 40.0 40.0 \n", |
||
| 1504 | "1 ID_4a8d7ec19f 30.0 NaN NaN \n", |
||
| 1505 | "2 ID_a6ca244172 30.0 NaN NaN \n", |
||
| 1506 | "3 ID_fa950a03af 30.0 NaN NaN \n", |
||
| 1507 | "4 ID_965d8b3d8e NaN 36.0 36.0 \n", |
||
| 1508 | "\n", |
||
| 1509 | " WindowWidth WindowWidth_0 WindowWidth_1 \n", |
||
| 1510 | "0 NaN 80.0 80.0 \n", |
||
| 1511 | "1 80.0 NaN NaN \n", |
||
| 1512 | "2 80.0 NaN NaN \n", |
||
| 1513 | "3 80.0 NaN NaN \n", |
||
| 1514 | "4 NaN 80.0 80.0 \n", |
||
| 1515 | "\n", |
||
| 1516 | "[5 rows x 33 columns]" |
||
| 1517 | ] |
||
| 1518 | }, |
||
| 1519 | "execution_count": 45, |
||
| 1520 | "metadata": {}, |
||
| 1521 | "output_type": "execute_result" |
||
| 1522 | } |
||
| 1523 | ], |
||
| 1524 | "source": [ |
||
| 1525 | "df_test_dicom = pd.DataFrame(l)\n", |
||
| 1526 | "df_test_dicom.head()" |
||
| 1527 | ] |
||
| 1528 | }, |
||
| 1529 | { |
||
| 1530 | "cell_type": "code", |
||
| 1531 | "execution_count": 46, |
||
| 1532 | "metadata": {}, |
||
| 1533 | "outputs": [], |
||
| 1534 | "source": [ |
||
| 1535 | "df_test_dicom.to_csv(f'data/{stage}_test_dicom.csv')" |
||
| 1536 | ] |
||
| 1537 | }, |
||
| 1538 | { |
||
| 1539 | "cell_type": "markdown", |
||
| 1540 | "metadata": {}, |
||
| 1541 | "source": [ |
||
| 1542 | "# Add labels to the train dicom csv\n", |
||
| 1543 | "Generates:\n", |
||
| 1544 | "* `train_dicom_diags.csv` (previously named `train_dicom_pivot.csv`)" |
||
| 1545 | ] |
||
| 1546 | }, |
||
| 1547 | { |
||
| 1548 | "cell_type": "code", |
||
| 1549 | "execution_count": 47, |
||
| 1550 | "metadata": {}, |
||
| 1551 | "outputs": [], |
||
| 1552 | "source": [ |
||
| 1553 | "df_train_dicom_diags = pd.merge(df_train_dicom, df_diags, how='left', left_on=['SOPInstanceUID'], right_on = ['fid'])" |
||
| 1554 | ] |
||
| 1555 | }, |
||
| 1556 | { |
||
| 1557 | "cell_type": "code", |
||
| 1558 | "execution_count": 48, |
||
| 1559 | "metadata": {}, |
||
| 1560 | "outputs": [], |
||
| 1561 | "source": [ |
||
| 1562 | "assert len(df_train_dicom) == len(df_diags) == len(df_train_dicom_diags)" |
||
| 1563 | ] |
||
| 1564 | }, |
||
| 1565 | { |
||
| 1566 | "cell_type": "code", |
||
| 1567 | "execution_count": 49, |
||
| 1568 | "metadata": {}, |
||
| 1569 | "outputs": [], |
||
| 1570 | "source": [ |
||
| 1571 | "df_train_dicom_diags.to_csv(f'data/{stage}_train_dicom_diags.csv')" |
||
| 1572 | ] |
||
| 1573 | } |
||
| 1574 | ], |
||
| 1575 | "metadata": { |
||
| 1576 | "kernelspec": { |
||
| 1577 | "display_name": "Python 3", |
||
| 1578 | "language": "python", |
||
| 1579 | "name": "python3" |
||
| 1580 | }, |
||
| 1581 | "language_info": { |
||
| 1582 | "codemirror_mode": { |
||
| 1583 | "name": "ipython", |
||
| 1584 | "version": 3 |
||
| 1585 | }, |
||
| 1586 | "file_extension": ".py", |
||
| 1587 | "mimetype": "text/x-python", |
||
| 1588 | "name": "python", |
||
| 1589 | "nbconvert_exporter": "python", |
||
| 1590 | "pygments_lexer": "ipython3", |
||
| 1591 | "version": "3.7.3" |
||
| 1592 | } |
||
| 1593 | }, |
||
| 1594 | "nbformat": 4, |
||
| 1595 | "nbformat_minor": 2 |
||
| 1596 | } |