Datasets
Models
Downloads: 1
Diff of
/2-preprocess-pickle.ipynb
[000000]
..
[fc9ccf]
Switch to unified view
| a | b/2-preprocess-pickle.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 collections import defaultdict\n", |
||
| 12 | "import pickle" |
||
| 13 | ] |
||
| 14 | }, |
||
| 15 | { |
||
| 16 | "cell_type": "code", |
||
| 17 | "execution_count": 2, |
||
| 18 | "metadata": {}, |
||
| 19 | "outputs": [], |
||
| 20 | "source": [ |
||
| 21 | "stage = \"stage_2\"" |
||
| 22 | ] |
||
| 23 | }, |
||
| 24 | { |
||
| 25 | "cell_type": "markdown", |
||
| 26 | "metadata": {}, |
||
| 27 | "source": [ |
||
| 28 | "# Train dataset" |
||
| 29 | ] |
||
| 30 | }, |
||
| 31 | { |
||
| 32 | "cell_type": "code", |
||
| 33 | "execution_count": 3, |
||
| 34 | "metadata": {}, |
||
| 35 | "outputs": [], |
||
| 36 | "source": [ |
||
| 37 | "df = pd.read_csv(f\"data/{stage}_train_dicom_diags.csv\")" |
||
| 38 | ] |
||
| 39 | }, |
||
| 40 | { |
||
| 41 | "cell_type": "code", |
||
| 42 | "execution_count": 4, |
||
| 43 | "metadata": {}, |
||
| 44 | "outputs": [ |
||
| 45 | { |
||
| 46 | "data": { |
||
| 47 | "text/html": [ |
||
| 48 | "<div>\n", |
||
| 49 | "<style scoped>\n", |
||
| 50 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 51 | " vertical-align: middle;\n", |
||
| 52 | " }\n", |
||
| 53 | "\n", |
||
| 54 | " .dataframe tbody tr th {\n", |
||
| 55 | " vertical-align: top;\n", |
||
| 56 | " }\n", |
||
| 57 | "\n", |
||
| 58 | " .dataframe thead th {\n", |
||
| 59 | " text-align: right;\n", |
||
| 60 | " }\n", |
||
| 61 | "</style>\n", |
||
| 62 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 63 | " <thead>\n", |
||
| 64 | " <tr style=\"text-align: right;\">\n", |
||
| 65 | " <th></th>\n", |
||
| 66 | " <th>Unnamed: 0</th>\n", |
||
| 67 | " <th>BitsAllocated</th>\n", |
||
| 68 | " <th>BitsStored</th>\n", |
||
| 69 | " <th>Columns</th>\n", |
||
| 70 | " <th>HighBit</th>\n", |
||
| 71 | " <th>ImageOrientationPatient_0</th>\n", |
||
| 72 | " <th>ImageOrientationPatient_1</th>\n", |
||
| 73 | " <th>ImageOrientationPatient_2</th>\n", |
||
| 74 | " <th>ImageOrientationPatient_3</th>\n", |
||
| 75 | " <th>ImageOrientationPatient_4</th>\n", |
||
| 76 | " <th>...</th>\n", |
||
| 77 | " <th>WindowWidth</th>\n", |
||
| 78 | " <th>WindowWidth_0</th>\n", |
||
| 79 | " <th>WindowWidth_1</th>\n", |
||
| 80 | " <th>fid</th>\n", |
||
| 81 | " <th>any</th>\n", |
||
| 82 | " <th>epidural</th>\n", |
||
| 83 | " <th>intraparenchymal</th>\n", |
||
| 84 | " <th>intraventricular</th>\n", |
||
| 85 | " <th>subarachnoid</th>\n", |
||
| 86 | " <th>subdural</th>\n", |
||
| 87 | " </tr>\n", |
||
| 88 | " </thead>\n", |
||
| 89 | " <tbody>\n", |
||
| 90 | " <tr>\n", |
||
| 91 | " <th>0</th>\n", |
||
| 92 | " <td>0</td>\n", |
||
| 93 | " <td>16</td>\n", |
||
| 94 | " <td>16</td>\n", |
||
| 95 | " <td>512</td>\n", |
||
| 96 | " <td>15</td>\n", |
||
| 97 | " <td>1.0</td>\n", |
||
| 98 | " <td>0.0</td>\n", |
||
| 99 | " <td>0.0</td>\n", |
||
| 100 | " <td>0.0</td>\n", |
||
| 101 | " <td>0.927184</td>\n", |
||
| 102 | " <td>...</td>\n", |
||
| 103 | " <td>80.0</td>\n", |
||
| 104 | " <td>NaN</td>\n", |
||
| 105 | " <td>NaN</td>\n", |
||
| 106 | " <td>ID_000012eaf</td>\n", |
||
| 107 | " <td>0</td>\n", |
||
| 108 | " <td>0</td>\n", |
||
| 109 | " <td>0</td>\n", |
||
| 110 | " <td>0</td>\n", |
||
| 111 | " <td>0</td>\n", |
||
| 112 | " <td>0</td>\n", |
||
| 113 | " </tr>\n", |
||
| 114 | " <tr>\n", |
||
| 115 | " <th>1</th>\n", |
||
| 116 | " <td>1</td>\n", |
||
| 117 | " <td>16</td>\n", |
||
| 118 | " <td>16</td>\n", |
||
| 119 | " <td>512</td>\n", |
||
| 120 | " <td>15</td>\n", |
||
| 121 | " <td>1.0</td>\n", |
||
| 122 | " <td>0.0</td>\n", |
||
| 123 | " <td>0.0</td>\n", |
||
| 124 | " <td>0.0</td>\n", |
||
| 125 | " <td>0.968148</td>\n", |
||
| 126 | " <td>...</td>\n", |
||
| 127 | " <td>80.0</td>\n", |
||
| 128 | " <td>NaN</td>\n", |
||
| 129 | " <td>NaN</td>\n", |
||
| 130 | " <td>ID_000039fa0</td>\n", |
||
| 131 | " <td>0</td>\n", |
||
| 132 | " <td>0</td>\n", |
||
| 133 | " <td>0</td>\n", |
||
| 134 | " <td>0</td>\n", |
||
| 135 | " <td>0</td>\n", |
||
| 136 | " <td>0</td>\n", |
||
| 137 | " </tr>\n", |
||
| 138 | " <tr>\n", |
||
| 139 | " <th>2</th>\n", |
||
| 140 | " <td>2</td>\n", |
||
| 141 | " <td>16</td>\n", |
||
| 142 | " <td>16</td>\n", |
||
| 143 | " <td>512</td>\n", |
||
| 144 | " <td>15</td>\n", |
||
| 145 | " <td>1.0</td>\n", |
||
| 146 | " <td>0.0</td>\n", |
||
| 147 | " <td>0.0</td>\n", |
||
| 148 | " <td>0.0</td>\n", |
||
| 149 | " <td>1.000000</td>\n", |
||
| 150 | " <td>...</td>\n", |
||
| 151 | " <td>100.0</td>\n", |
||
| 152 | " <td>NaN</td>\n", |
||
| 153 | " <td>NaN</td>\n", |
||
| 154 | " <td>ID_00005679d</td>\n", |
||
| 155 | " <td>0</td>\n", |
||
| 156 | " <td>0</td>\n", |
||
| 157 | " <td>0</td>\n", |
||
| 158 | " <td>0</td>\n", |
||
| 159 | " <td>0</td>\n", |
||
| 160 | " <td>0</td>\n", |
||
| 161 | " </tr>\n", |
||
| 162 | " <tr>\n", |
||
| 163 | " <th>3</th>\n", |
||
| 164 | " <td>3</td>\n", |
||
| 165 | " <td>16</td>\n", |
||
| 166 | " <td>12</td>\n", |
||
| 167 | " <td>512</td>\n", |
||
| 168 | " <td>11</td>\n", |
||
| 169 | " <td>1.0</td>\n", |
||
| 170 | " <td>0.0</td>\n", |
||
| 171 | " <td>0.0</td>\n", |
||
| 172 | " <td>0.0</td>\n", |
||
| 173 | " <td>0.994522</td>\n", |
||
| 174 | " <td>...</td>\n", |
||
| 175 | " <td>NaN</td>\n", |
||
| 176 | " <td>80.0</td>\n", |
||
| 177 | " <td>80.0</td>\n", |
||
| 178 | " <td>ID_00008ce3c</td>\n", |
||
| 179 | " <td>0</td>\n", |
||
| 180 | " <td>0</td>\n", |
||
| 181 | " <td>0</td>\n", |
||
| 182 | " <td>0</td>\n", |
||
| 183 | " <td>0</td>\n", |
||
| 184 | " <td>0</td>\n", |
||
| 185 | " </tr>\n", |
||
| 186 | " <tr>\n", |
||
| 187 | " <th>4</th>\n", |
||
| 188 | " <td>4</td>\n", |
||
| 189 | " <td>16</td>\n", |
||
| 190 | " <td>16</td>\n", |
||
| 191 | " <td>512</td>\n", |
||
| 192 | " <td>15</td>\n", |
||
| 193 | " <td>1.0</td>\n", |
||
| 194 | " <td>0.0</td>\n", |
||
| 195 | " <td>0.0</td>\n", |
||
| 196 | " <td>0.0</td>\n", |
||
| 197 | " <td>1.000000</td>\n", |
||
| 198 | " <td>...</td>\n", |
||
| 199 | " <td>135.0</td>\n", |
||
| 200 | " <td>NaN</td>\n", |
||
| 201 | " <td>NaN</td>\n", |
||
| 202 | " <td>ID_0000950d7</td>\n", |
||
| 203 | " <td>0</td>\n", |
||
| 204 | " <td>0</td>\n", |
||
| 205 | " <td>0</td>\n", |
||
| 206 | " <td>0</td>\n", |
||
| 207 | " <td>0</td>\n", |
||
| 208 | " <td>0</td>\n", |
||
| 209 | " </tr>\n", |
||
| 210 | " </tbody>\n", |
||
| 211 | "</table>\n", |
||
| 212 | "<p>5 rows × 41 columns</p>\n", |
||
| 213 | "</div>" |
||
| 214 | ], |
||
| 215 | "text/plain": [ |
||
| 216 | " Unnamed: 0 BitsAllocated BitsStored Columns HighBit \\\n", |
||
| 217 | "0 0 16 16 512 15 \n", |
||
| 218 | "1 1 16 16 512 15 \n", |
||
| 219 | "2 2 16 16 512 15 \n", |
||
| 220 | "3 3 16 12 512 11 \n", |
||
| 221 | "4 4 16 16 512 15 \n", |
||
| 222 | "\n", |
||
| 223 | " ImageOrientationPatient_0 ImageOrientationPatient_1 \\\n", |
||
| 224 | "0 1.0 0.0 \n", |
||
| 225 | "1 1.0 0.0 \n", |
||
| 226 | "2 1.0 0.0 \n", |
||
| 227 | "3 1.0 0.0 \n", |
||
| 228 | "4 1.0 0.0 \n", |
||
| 229 | "\n", |
||
| 230 | " ImageOrientationPatient_2 ImageOrientationPatient_3 \\\n", |
||
| 231 | "0 0.0 0.0 \n", |
||
| 232 | "1 0.0 0.0 \n", |
||
| 233 | "2 0.0 0.0 \n", |
||
| 234 | "3 0.0 0.0 \n", |
||
| 235 | "4 0.0 0.0 \n", |
||
| 236 | "\n", |
||
| 237 | " ImageOrientationPatient_4 ... WindowWidth WindowWidth_0 WindowWidth_1 \\\n", |
||
| 238 | "0 0.927184 ... 80.0 NaN NaN \n", |
||
| 239 | "1 0.968148 ... 80.0 NaN NaN \n", |
||
| 240 | "2 1.000000 ... 100.0 NaN NaN \n", |
||
| 241 | "3 0.994522 ... NaN 80.0 80.0 \n", |
||
| 242 | "4 1.000000 ... 135.0 NaN NaN \n", |
||
| 243 | "\n", |
||
| 244 | " fid any epidural intraparenchymal intraventricular subarachnoid \\\n", |
||
| 245 | "0 ID_000012eaf 0 0 0 0 0 \n", |
||
| 246 | "1 ID_000039fa0 0 0 0 0 0 \n", |
||
| 247 | "2 ID_00005679d 0 0 0 0 0 \n", |
||
| 248 | "3 ID_00008ce3c 0 0 0 0 0 \n", |
||
| 249 | "4 ID_0000950d7 0 0 0 0 0 \n", |
||
| 250 | "\n", |
||
| 251 | " subdural \n", |
||
| 252 | "0 0 \n", |
||
| 253 | "1 0 \n", |
||
| 254 | "2 0 \n", |
||
| 255 | "3 0 \n", |
||
| 256 | "4 0 \n", |
||
| 257 | "\n", |
||
| 258 | "[5 rows x 41 columns]" |
||
| 259 | ] |
||
| 260 | }, |
||
| 261 | "execution_count": 4, |
||
| 262 | "metadata": {}, |
||
| 263 | "output_type": "execute_result" |
||
| 264 | } |
||
| 265 | ], |
||
| 266 | "source": [ |
||
| 267 | "df.head()" |
||
| 268 | ] |
||
| 269 | }, |
||
| 270 | { |
||
| 271 | "cell_type": "code", |
||
| 272 | "execution_count": 5, |
||
| 273 | "metadata": {}, |
||
| 274 | "outputs": [ |
||
| 275 | { |
||
| 276 | "data": { |
||
| 277 | "text/plain": [ |
||
| 278 | "21744" |
||
| 279 | ] |
||
| 280 | }, |
||
| 281 | "execution_count": 5, |
||
| 282 | "metadata": {}, |
||
| 283 | "output_type": "execute_result" |
||
| 284 | } |
||
| 285 | ], |
||
| 286 | "source": [ |
||
| 287 | "# sort, then group by (order is preserved within groups)\n", |
||
| 288 | "gs = df.sort_values('ImagePositionPatient_2').groupby('SeriesInstanceUID')\n", |
||
| 289 | "len(gs)" |
||
| 290 | ] |
||
| 291 | }, |
||
| 292 | { |
||
| 293 | "cell_type": "code", |
||
| 294 | "execution_count": 6, |
||
| 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>ImagePositionPatient_2</th>\n", |
||
| 319 | " <th>fid</th>\n", |
||
| 320 | " </tr>\n", |
||
| 321 | " </thead>\n", |
||
| 322 | " <tbody>\n", |
||
| 323 | " <tr>\n", |
||
| 324 | " <th>577964</th>\n", |
||
| 325 | " <td>193.542489</td>\n", |
||
| 326 | " <td>ID_c45659d3d</td>\n", |
||
| 327 | " </tr>\n", |
||
| 328 | " <tr>\n", |
||
| 329 | " <th>229790</th>\n", |
||
| 330 | " <td>198.214051</td>\n", |
||
| 331 | " <td>ID_4e0bdd2ba</td>\n", |
||
| 332 | " </tr>\n", |
||
| 333 | " <tr>\n", |
||
| 334 | " <th>22395</th>\n", |
||
| 335 | " <td>202.885613</td>\n", |
||
| 336 | " <td>ID_079945c27</td>\n", |
||
| 337 | " </tr>\n", |
||
| 338 | " <tr>\n", |
||
| 339 | " <th>746126</th>\n", |
||
| 340 | " <td>207.557174</td>\n", |
||
| 341 | " <td>ID_fdbfb2c17</td>\n", |
||
| 342 | " </tr>\n", |
||
| 343 | " <tr>\n", |
||
| 344 | " <th>253266</th>\n", |
||
| 345 | " <td>212.228736</td>\n", |
||
| 346 | " <td>ID_55f7bbbf2</td>\n", |
||
| 347 | " </tr>\n", |
||
| 348 | " </tbody>\n", |
||
| 349 | "</table>\n", |
||
| 350 | "</div>" |
||
| 351 | ], |
||
| 352 | "text/plain": [ |
||
| 353 | " ImagePositionPatient_2 fid\n", |
||
| 354 | "577964 193.542489 ID_c45659d3d\n", |
||
| 355 | "229790 198.214051 ID_4e0bdd2ba\n", |
||
| 356 | "22395 202.885613 ID_079945c27\n", |
||
| 357 | "746126 207.557174 ID_fdbfb2c17\n", |
||
| 358 | "253266 212.228736 ID_55f7bbbf2" |
||
| 359 | ] |
||
| 360 | }, |
||
| 361 | "execution_count": 6, |
||
| 362 | "metadata": {}, |
||
| 363 | "output_type": "execute_result" |
||
| 364 | } |
||
| 365 | ], |
||
| 366 | "source": [ |
||
| 367 | "# see if it worked\n", |
||
| 368 | "gs.get_group('ID_fa19cd5ea9')[['ImagePositionPatient_2', 'fid']].head()" |
||
| 369 | ] |
||
| 370 | }, |
||
| 371 | { |
||
| 372 | "cell_type": "code", |
||
| 373 | "execution_count": 7, |
||
| 374 | "metadata": {}, |
||
| 375 | "outputs": [], |
||
| 376 | "source": [ |
||
| 377 | "g = gs.get_group('ID_fa19cd5ea9')" |
||
| 378 | ] |
||
| 379 | }, |
||
| 380 | { |
||
| 381 | "cell_type": "code", |
||
| 382 | "execution_count": 8, |
||
| 383 | "metadata": {}, |
||
| 384 | "outputs": [], |
||
| 385 | "source": [ |
||
| 386 | "subg = g[['SeriesInstanceUID', 'fid', 'any', 'epidural', \n", |
||
| 387 | " 'intraparenchymal', 'intraventricular', 'subarachnoid', 'subdural']]" |
||
| 388 | ] |
||
| 389 | }, |
||
| 390 | { |
||
| 391 | "cell_type": "code", |
||
| 392 | "execution_count": 9, |
||
| 393 | "metadata": {}, |
||
| 394 | "outputs": [ |
||
| 395 | { |
||
| 396 | "data": { |
||
| 397 | "text/html": [ |
||
| 398 | "<div>\n", |
||
| 399 | "<style scoped>\n", |
||
| 400 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 401 | " vertical-align: middle;\n", |
||
| 402 | " }\n", |
||
| 403 | "\n", |
||
| 404 | " .dataframe tbody tr th {\n", |
||
| 405 | " vertical-align: top;\n", |
||
| 406 | " }\n", |
||
| 407 | "\n", |
||
| 408 | " .dataframe thead th {\n", |
||
| 409 | " text-align: right;\n", |
||
| 410 | " }\n", |
||
| 411 | "</style>\n", |
||
| 412 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 413 | " <thead>\n", |
||
| 414 | " <tr style=\"text-align: right;\">\n", |
||
| 415 | " <th></th>\n", |
||
| 416 | " <th>SeriesInstanceUID</th>\n", |
||
| 417 | " <th>fid</th>\n", |
||
| 418 | " <th>any</th>\n", |
||
| 419 | " <th>epidural</th>\n", |
||
| 420 | " <th>intraparenchymal</th>\n", |
||
| 421 | " <th>intraventricular</th>\n", |
||
| 422 | " <th>subarachnoid</th>\n", |
||
| 423 | " <th>subdural</th>\n", |
||
| 424 | " </tr>\n", |
||
| 425 | " </thead>\n", |
||
| 426 | " <tbody>\n", |
||
| 427 | " <tr>\n", |
||
| 428 | " <th>577964</th>\n", |
||
| 429 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 430 | " <td>ID_c45659d3d</td>\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>229790</th>\n", |
||
| 440 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 441 | " <td>ID_4e0bdd2ba</td>\n", |
||
| 442 | " <td>0</td>\n", |
||
| 443 | " <td>0</td>\n", |
||
| 444 | " <td>0</td>\n", |
||
| 445 | " <td>0</td>\n", |
||
| 446 | " <td>0</td>\n", |
||
| 447 | " <td>0</td>\n", |
||
| 448 | " </tr>\n", |
||
| 449 | " <tr>\n", |
||
| 450 | " <th>22395</th>\n", |
||
| 451 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 452 | " <td>ID_079945c27</td>\n", |
||
| 453 | " <td>1</td>\n", |
||
| 454 | " <td>0</td>\n", |
||
| 455 | " <td>0</td>\n", |
||
| 456 | " <td>0</td>\n", |
||
| 457 | " <td>0</td>\n", |
||
| 458 | " <td>1</td>\n", |
||
| 459 | " </tr>\n", |
||
| 460 | " <tr>\n", |
||
| 461 | " <th>746126</th>\n", |
||
| 462 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 463 | " <td>ID_fdbfb2c17</td>\n", |
||
| 464 | " <td>1</td>\n", |
||
| 465 | " <td>0</td>\n", |
||
| 466 | " <td>0</td>\n", |
||
| 467 | " <td>0</td>\n", |
||
| 468 | " <td>0</td>\n", |
||
| 469 | " <td>1</td>\n", |
||
| 470 | " </tr>\n", |
||
| 471 | " <tr>\n", |
||
| 472 | " <th>253266</th>\n", |
||
| 473 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 474 | " <td>ID_55f7bbbf2</td>\n", |
||
| 475 | " <td>1</td>\n", |
||
| 476 | " <td>0</td>\n", |
||
| 477 | " <td>0</td>\n", |
||
| 478 | " <td>0</td>\n", |
||
| 479 | " <td>0</td>\n", |
||
| 480 | " <td>1</td>\n", |
||
| 481 | " </tr>\n", |
||
| 482 | " <tr>\n", |
||
| 483 | " <th>549211</th>\n", |
||
| 484 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 485 | " <td>ID_ba7080372</td>\n", |
||
| 486 | " <td>1</td>\n", |
||
| 487 | " <td>0</td>\n", |
||
| 488 | " <td>0</td>\n", |
||
| 489 | " <td>0</td>\n", |
||
| 490 | " <td>0</td>\n", |
||
| 491 | " <td>1</td>\n", |
||
| 492 | " </tr>\n", |
||
| 493 | " <tr>\n", |
||
| 494 | " <th>592856</th>\n", |
||
| 495 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 496 | " <td>ID_c964e4096</td>\n", |
||
| 497 | " <td>1</td>\n", |
||
| 498 | " <td>0</td>\n", |
||
| 499 | " <td>0</td>\n", |
||
| 500 | " <td>0</td>\n", |
||
| 501 | " <td>0</td>\n", |
||
| 502 | " <td>1</td>\n", |
||
| 503 | " </tr>\n", |
||
| 504 | " <tr>\n", |
||
| 505 | " <th>183149</th>\n", |
||
| 506 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 507 | " <td>ID_3e31d57d0</td>\n", |
||
| 508 | " <td>1</td>\n", |
||
| 509 | " <td>0</td>\n", |
||
| 510 | " <td>0</td>\n", |
||
| 511 | " <td>0</td>\n", |
||
| 512 | " <td>0</td>\n", |
||
| 513 | " <td>1</td>\n", |
||
| 514 | " </tr>\n", |
||
| 515 | " <tr>\n", |
||
| 516 | " <th>306771</th>\n", |
||
| 517 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 518 | " <td>ID_680b2194c</td>\n", |
||
| 519 | " <td>1</td>\n", |
||
| 520 | " <td>0</td>\n", |
||
| 521 | " <td>0</td>\n", |
||
| 522 | " <td>0</td>\n", |
||
| 523 | " <td>0</td>\n", |
||
| 524 | " <td>1</td>\n", |
||
| 525 | " </tr>\n", |
||
| 526 | " <tr>\n", |
||
| 527 | " <th>540358</th>\n", |
||
| 528 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 529 | " <td>ID_b76b13444</td>\n", |
||
| 530 | " <td>1</td>\n", |
||
| 531 | " <td>0</td>\n", |
||
| 532 | " <td>0</td>\n", |
||
| 533 | " <td>0</td>\n", |
||
| 534 | " <td>0</td>\n", |
||
| 535 | " <td>1</td>\n", |
||
| 536 | " </tr>\n", |
||
| 537 | " <tr>\n", |
||
| 538 | " <th>645217</th>\n", |
||
| 539 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 540 | " <td>ID_db48a633d</td>\n", |
||
| 541 | " <td>1</td>\n", |
||
| 542 | " <td>0</td>\n", |
||
| 543 | " <td>0</td>\n", |
||
| 544 | " <td>0</td>\n", |
||
| 545 | " <td>0</td>\n", |
||
| 546 | " <td>1</td>\n", |
||
| 547 | " </tr>\n", |
||
| 548 | " <tr>\n", |
||
| 549 | " <th>270974</th>\n", |
||
| 550 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 551 | " <td>ID_5bf2ca43f</td>\n", |
||
| 552 | " <td>1</td>\n", |
||
| 553 | " <td>0</td>\n", |
||
| 554 | " <td>0</td>\n", |
||
| 555 | " <td>0</td>\n", |
||
| 556 | " <td>0</td>\n", |
||
| 557 | " <td>1</td>\n", |
||
| 558 | " </tr>\n", |
||
| 559 | " <tr>\n", |
||
| 560 | " <th>672814</th>\n", |
||
| 561 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 562 | " <td>ID_e4b636907</td>\n", |
||
| 563 | " <td>1</td>\n", |
||
| 564 | " <td>0</td>\n", |
||
| 565 | " <td>0</td>\n", |
||
| 566 | " <td>0</td>\n", |
||
| 567 | " <td>0</td>\n", |
||
| 568 | " <td>1</td>\n", |
||
| 569 | " </tr>\n", |
||
| 570 | " <tr>\n", |
||
| 571 | " <th>350834</th>\n", |
||
| 572 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 573 | " <td>ID_7714ead69</td>\n", |
||
| 574 | " <td>1</td>\n", |
||
| 575 | " <td>0</td>\n", |
||
| 576 | " <td>0</td>\n", |
||
| 577 | " <td>0</td>\n", |
||
| 578 | " <td>0</td>\n", |
||
| 579 | " <td>1</td>\n", |
||
| 580 | " </tr>\n", |
||
| 581 | " <tr>\n", |
||
| 582 | " <th>749886</th>\n", |
||
| 583 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 584 | " <td>ID_ff012ee5b</td>\n", |
||
| 585 | " <td>1</td>\n", |
||
| 586 | " <td>0</td>\n", |
||
| 587 | " <td>0</td>\n", |
||
| 588 | " <td>0</td>\n", |
||
| 589 | " <td>0</td>\n", |
||
| 590 | " <td>1</td>\n", |
||
| 591 | " </tr>\n", |
||
| 592 | " <tr>\n", |
||
| 593 | " <th>523978</th>\n", |
||
| 594 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 595 | " <td>ID_b1cea5abb</td>\n", |
||
| 596 | " <td>1</td>\n", |
||
| 597 | " <td>0</td>\n", |
||
| 598 | " <td>0</td>\n", |
||
| 599 | " <td>0</td>\n", |
||
| 600 | " <td>0</td>\n", |
||
| 601 | " <td>1</td>\n", |
||
| 602 | " </tr>\n", |
||
| 603 | " <tr>\n", |
||
| 604 | " <th>464942</th>\n", |
||
| 605 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 606 | " <td>ID_9dad2eb09</td>\n", |
||
| 607 | " <td>1</td>\n", |
||
| 608 | " <td>0</td>\n", |
||
| 609 | " <td>0</td>\n", |
||
| 610 | " <td>0</td>\n", |
||
| 611 | " <td>0</td>\n", |
||
| 612 | " <td>1</td>\n", |
||
| 613 | " </tr>\n", |
||
| 614 | " <tr>\n", |
||
| 615 | " <th>229881</th>\n", |
||
| 616 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 617 | " <td>ID_4e14d0fe8</td>\n", |
||
| 618 | " <td>1</td>\n", |
||
| 619 | " <td>0</td>\n", |
||
| 620 | " <td>0</td>\n", |
||
| 621 | " <td>0</td>\n", |
||
| 622 | " <td>0</td>\n", |
||
| 623 | " <td>1</td>\n", |
||
| 624 | " </tr>\n", |
||
| 625 | " <tr>\n", |
||
| 626 | " <th>186237</th>\n", |
||
| 627 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 628 | " <td>ID_3f422852d</td>\n", |
||
| 629 | " <td>1</td>\n", |
||
| 630 | " <td>0</td>\n", |
||
| 631 | " <td>0</td>\n", |
||
| 632 | " <td>0</td>\n", |
||
| 633 | " <td>0</td>\n", |
||
| 634 | " <td>1</td>\n", |
||
| 635 | " </tr>\n", |
||
| 636 | " <tr>\n", |
||
| 637 | " <th>599624</th>\n", |
||
| 638 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 639 | " <td>ID_cbbb50e6d</td>\n", |
||
| 640 | " <td>1</td>\n", |
||
| 641 | " <td>0</td>\n", |
||
| 642 | " <td>0</td>\n", |
||
| 643 | " <td>0</td>\n", |
||
| 644 | " <td>0</td>\n", |
||
| 645 | " <td>1</td>\n", |
||
| 646 | " </tr>\n", |
||
| 647 | " <tr>\n", |
||
| 648 | " <th>347055</th>\n", |
||
| 649 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 650 | " <td>ID_75cbdae68</td>\n", |
||
| 651 | " <td>1</td>\n", |
||
| 652 | " <td>0</td>\n", |
||
| 653 | " <td>0</td>\n", |
||
| 654 | " <td>0</td>\n", |
||
| 655 | " <td>0</td>\n", |
||
| 656 | " <td>1</td>\n", |
||
| 657 | " </tr>\n", |
||
| 658 | " <tr>\n", |
||
| 659 | " <th>359450</th>\n", |
||
| 660 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 661 | " <td>ID_7a02fdbea</td>\n", |
||
| 662 | " <td>1</td>\n", |
||
| 663 | " <td>0</td>\n", |
||
| 664 | " <td>0</td>\n", |
||
| 665 | " <td>0</td>\n", |
||
| 666 | " <td>0</td>\n", |
||
| 667 | " <td>1</td>\n", |
||
| 668 | " </tr>\n", |
||
| 669 | " <tr>\n", |
||
| 670 | " <th>127205</th>\n", |
||
| 671 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 672 | " <td>ID_2b3671dd9</td>\n", |
||
| 673 | " <td>1</td>\n", |
||
| 674 | " <td>0</td>\n", |
||
| 675 | " <td>0</td>\n", |
||
| 676 | " <td>0</td>\n", |
||
| 677 | " <td>0</td>\n", |
||
| 678 | " <td>1</td>\n", |
||
| 679 | " </tr>\n", |
||
| 680 | " <tr>\n", |
||
| 681 | " <th>148587</th>\n", |
||
| 682 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 683 | " <td>ID_3274f5977</td>\n", |
||
| 684 | " <td>0</td>\n", |
||
| 685 | " <td>0</td>\n", |
||
| 686 | " <td>0</td>\n", |
||
| 687 | " <td>0</td>\n", |
||
| 688 | " <td>0</td>\n", |
||
| 689 | " <td>0</td>\n", |
||
| 690 | " </tr>\n", |
||
| 691 | " <tr>\n", |
||
| 692 | " <th>413641</th>\n", |
||
| 693 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 694 | " <td>ID_8c5fc9e44</td>\n", |
||
| 695 | " <td>0</td>\n", |
||
| 696 | " <td>0</td>\n", |
||
| 697 | " <td>0</td>\n", |
||
| 698 | " <td>0</td>\n", |
||
| 699 | " <td>0</td>\n", |
||
| 700 | " <td>0</td>\n", |
||
| 701 | " </tr>\n", |
||
| 702 | " <tr>\n", |
||
| 703 | " <th>688538</th>\n", |
||
| 704 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 705 | " <td>ID_ea2861e9a</td>\n", |
||
| 706 | " <td>0</td>\n", |
||
| 707 | " <td>0</td>\n", |
||
| 708 | " <td>0</td>\n", |
||
| 709 | " <td>0</td>\n", |
||
| 710 | " <td>0</td>\n", |
||
| 711 | " <td>0</td>\n", |
||
| 712 | " </tr>\n", |
||
| 713 | " <tr>\n", |
||
| 714 | " <th>318670</th>\n", |
||
| 715 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 716 | " <td>ID_6c19c9f7b</td>\n", |
||
| 717 | " <td>0</td>\n", |
||
| 718 | " <td>0</td>\n", |
||
| 719 | " <td>0</td>\n", |
||
| 720 | " <td>0</td>\n", |
||
| 721 | " <td>0</td>\n", |
||
| 722 | " <td>0</td>\n", |
||
| 723 | " </tr>\n", |
||
| 724 | " <tr>\n", |
||
| 725 | " <th>630472</th>\n", |
||
| 726 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 727 | " <td>ID_d6435f3bf</td>\n", |
||
| 728 | " <td>0</td>\n", |
||
| 729 | " <td>0</td>\n", |
||
| 730 | " <td>0</td>\n", |
||
| 731 | " <td>0</td>\n", |
||
| 732 | " <td>0</td>\n", |
||
| 733 | " <td>0</td>\n", |
||
| 734 | " </tr>\n", |
||
| 735 | " <tr>\n", |
||
| 736 | " <th>202656</th>\n", |
||
| 737 | " <td>ID_fa19cd5ea9</td>\n", |
||
| 738 | " <td>ID_44d57858e</td>\n", |
||
| 739 | " <td>0</td>\n", |
||
| 740 | " <td>0</td>\n", |
||
| 741 | " <td>0</td>\n", |
||
| 742 | " <td>0</td>\n", |
||
| 743 | " <td>0</td>\n", |
||
| 744 | " <td>0</td>\n", |
||
| 745 | " </tr>\n", |
||
| 746 | " </tbody>\n", |
||
| 747 | "</table>\n", |
||
| 748 | "</div>" |
||
| 749 | ], |
||
| 750 | "text/plain": [ |
||
| 751 | " SeriesInstanceUID fid any epidural intraparenchymal \\\n", |
||
| 752 | "577964 ID_fa19cd5ea9 ID_c45659d3d 0 0 0 \n", |
||
| 753 | "229790 ID_fa19cd5ea9 ID_4e0bdd2ba 0 0 0 \n", |
||
| 754 | "22395 ID_fa19cd5ea9 ID_079945c27 1 0 0 \n", |
||
| 755 | "746126 ID_fa19cd5ea9 ID_fdbfb2c17 1 0 0 \n", |
||
| 756 | "253266 ID_fa19cd5ea9 ID_55f7bbbf2 1 0 0 \n", |
||
| 757 | "549211 ID_fa19cd5ea9 ID_ba7080372 1 0 0 \n", |
||
| 758 | "592856 ID_fa19cd5ea9 ID_c964e4096 1 0 0 \n", |
||
| 759 | "183149 ID_fa19cd5ea9 ID_3e31d57d0 1 0 0 \n", |
||
| 760 | "306771 ID_fa19cd5ea9 ID_680b2194c 1 0 0 \n", |
||
| 761 | "540358 ID_fa19cd5ea9 ID_b76b13444 1 0 0 \n", |
||
| 762 | "645217 ID_fa19cd5ea9 ID_db48a633d 1 0 0 \n", |
||
| 763 | "270974 ID_fa19cd5ea9 ID_5bf2ca43f 1 0 0 \n", |
||
| 764 | "672814 ID_fa19cd5ea9 ID_e4b636907 1 0 0 \n", |
||
| 765 | "350834 ID_fa19cd5ea9 ID_7714ead69 1 0 0 \n", |
||
| 766 | "749886 ID_fa19cd5ea9 ID_ff012ee5b 1 0 0 \n", |
||
| 767 | "523978 ID_fa19cd5ea9 ID_b1cea5abb 1 0 0 \n", |
||
| 768 | "464942 ID_fa19cd5ea9 ID_9dad2eb09 1 0 0 \n", |
||
| 769 | "229881 ID_fa19cd5ea9 ID_4e14d0fe8 1 0 0 \n", |
||
| 770 | "186237 ID_fa19cd5ea9 ID_3f422852d 1 0 0 \n", |
||
| 771 | "599624 ID_fa19cd5ea9 ID_cbbb50e6d 1 0 0 \n", |
||
| 772 | "347055 ID_fa19cd5ea9 ID_75cbdae68 1 0 0 \n", |
||
| 773 | "359450 ID_fa19cd5ea9 ID_7a02fdbea 1 0 0 \n", |
||
| 774 | "127205 ID_fa19cd5ea9 ID_2b3671dd9 1 0 0 \n", |
||
| 775 | "148587 ID_fa19cd5ea9 ID_3274f5977 0 0 0 \n", |
||
| 776 | "413641 ID_fa19cd5ea9 ID_8c5fc9e44 0 0 0 \n", |
||
| 777 | "688538 ID_fa19cd5ea9 ID_ea2861e9a 0 0 0 \n", |
||
| 778 | "318670 ID_fa19cd5ea9 ID_6c19c9f7b 0 0 0 \n", |
||
| 779 | "630472 ID_fa19cd5ea9 ID_d6435f3bf 0 0 0 \n", |
||
| 780 | "202656 ID_fa19cd5ea9 ID_44d57858e 0 0 0 \n", |
||
| 781 | "\n", |
||
| 782 | " intraventricular subarachnoid subdural \n", |
||
| 783 | "577964 0 0 0 \n", |
||
| 784 | "229790 0 0 0 \n", |
||
| 785 | "22395 0 0 1 \n", |
||
| 786 | "746126 0 0 1 \n", |
||
| 787 | "253266 0 0 1 \n", |
||
| 788 | "549211 0 0 1 \n", |
||
| 789 | "592856 0 0 1 \n", |
||
| 790 | "183149 0 0 1 \n", |
||
| 791 | "306771 0 0 1 \n", |
||
| 792 | "540358 0 0 1 \n", |
||
| 793 | "645217 0 0 1 \n", |
||
| 794 | "270974 0 0 1 \n", |
||
| 795 | "672814 0 0 1 \n", |
||
| 796 | "350834 0 0 1 \n", |
||
| 797 | "749886 0 0 1 \n", |
||
| 798 | "523978 0 0 1 \n", |
||
| 799 | "464942 0 0 1 \n", |
||
| 800 | "229881 0 0 1 \n", |
||
| 801 | "186237 0 0 1 \n", |
||
| 802 | "599624 0 0 1 \n", |
||
| 803 | "347055 0 0 1 \n", |
||
| 804 | "359450 0 0 1 \n", |
||
| 805 | "127205 0 0 1 \n", |
||
| 806 | "148587 0 0 0 \n", |
||
| 807 | "413641 0 0 0 \n", |
||
| 808 | "688538 0 0 0 \n", |
||
| 809 | "318670 0 0 0 \n", |
||
| 810 | "630472 0 0 0 \n", |
||
| 811 | "202656 0 0 0 " |
||
| 812 | ] |
||
| 813 | }, |
||
| 814 | "execution_count": 9, |
||
| 815 | "metadata": {}, |
||
| 816 | "output_type": "execute_result" |
||
| 817 | } |
||
| 818 | ], |
||
| 819 | "source": [ |
||
| 820 | "subg" |
||
| 821 | ] |
||
| 822 | }, |
||
| 823 | { |
||
| 824 | "cell_type": "code", |
||
| 825 | "execution_count": 10, |
||
| 826 | "metadata": {}, |
||
| 827 | "outputs": [], |
||
| 828 | "source": [ |
||
| 829 | "# You can use a btrfs snapshot and rename files by study_id and z-pos through the brain\n", |
||
| 830 | "def rename_train_group(subg):\n", |
||
| 831 | " ix = 0\n", |
||
| 832 | " total = len(subg)\n", |
||
| 833 | " for index, row in subg.iterrows():\n", |
||
| 834 | " cur_fn = row['fid']\n", |
||
| 835 | " new_fn = f\"{row['SeriesInstanceUID']}_{ix:03}_{total:03}_{row['any']}_{row['epidural']}_{row['intraparenchymal']}_{row['intraventricular']}_{row['subarachnoid']}_{row['subdural']}_{cur_fn}\"\n", |
||
| 836 | " ix += 1\n", |
||
| 837 | " Path(f'data/unzip_renamed/{stage}_train_images/{cur_fn}.dcm').rename(f'data/unzip_renamed/{stage}_train_images/{new_fn}.dcm')" |
||
| 838 | ] |
||
| 839 | }, |
||
| 840 | { |
||
| 841 | "cell_type": "code", |
||
| 842 | "execution_count": 11, |
||
| 843 | "metadata": {}, |
||
| 844 | "outputs": [], |
||
| 845 | "source": [ |
||
| 846 | "def index_group(subg, study_ix_to_fn, fn_to_study_ix):\n", |
||
| 847 | " ix = 0\n", |
||
| 848 | " for index, row in subg.iterrows():\n", |
||
| 849 | " fn = row['SOPInstanceUID']\n", |
||
| 850 | " study = row['SeriesInstanceUID']\n", |
||
| 851 | " study_ix_to_fn[study].append(fn)\n", |
||
| 852 | " fn_to_study_ix[fn] = (study, ix)\n", |
||
| 853 | " ix += 1 " |
||
| 854 | ] |
||
| 855 | }, |
||
| 856 | { |
||
| 857 | "cell_type": "code", |
||
| 858 | "execution_count": 12, |
||
| 859 | "metadata": {}, |
||
| 860 | "outputs": [], |
||
| 861 | "source": [ |
||
| 862 | "labels = [ 'any', 'epidural', 'intraparenchymal', 'intraventricular', 'subarachnoid', 'subdural' ]\n", |
||
| 863 | "\n", |
||
| 864 | "def label_group(subg, fn_to_labels):\n", |
||
| 865 | " for index, row in subg.iterrows():\n", |
||
| 866 | " fn = row['SOPInstanceUID']\n", |
||
| 867 | " fn_to_labels[fn] = [ label for label in labels if row[label] == 1 ]" |
||
| 868 | ] |
||
| 869 | }, |
||
| 870 | { |
||
| 871 | "cell_type": "code", |
||
| 872 | "execution_count": 13, |
||
| 873 | "metadata": {}, |
||
| 874 | "outputs": [], |
||
| 875 | "source": [ |
||
| 876 | "train_study_ix_to_fn = defaultdict(list)\n", |
||
| 877 | "train_fn_to_study_ix = {}\n", |
||
| 878 | "train_fn_to_labels = {}\n", |
||
| 879 | "\n", |
||
| 880 | "for name, subg in gs:\n", |
||
| 881 | " #rename_train_group(subg)\n", |
||
| 882 | " index_group(subg, train_study_ix_to_fn, train_fn_to_study_ix)\n", |
||
| 883 | " label_group(subg, train_fn_to_labels)" |
||
| 884 | ] |
||
| 885 | }, |
||
| 886 | { |
||
| 887 | "cell_type": "code", |
||
| 888 | "execution_count": 14, |
||
| 889 | "metadata": {}, |
||
| 890 | "outputs": [], |
||
| 891 | "source": [ |
||
| 892 | "# Do not pickle yet, merge with test\n", |
||
| 893 | "pickle.dump(train_study_ix_to_fn, open(f\"data/{stage}_train_study_ix_to_fn.pickle\", \"wb\" ))\n", |
||
| 894 | "pickle.dump(train_fn_to_study_ix, open(f\"data/{stage}_train_fn_to_study_ix.pickle\", \"wb\" ))" |
||
| 895 | ] |
||
| 896 | }, |
||
| 897 | { |
||
| 898 | "cell_type": "code", |
||
| 899 | "execution_count": 15, |
||
| 900 | "metadata": {}, |
||
| 901 | "outputs": [ |
||
| 902 | { |
||
| 903 | "data": { |
||
| 904 | "text/plain": [ |
||
| 905 | "['any', 'subdural']" |
||
| 906 | ] |
||
| 907 | }, |
||
| 908 | "execution_count": 15, |
||
| 909 | "metadata": {}, |
||
| 910 | "output_type": "execute_result" |
||
| 911 | } |
||
| 912 | ], |
||
| 913 | "source": [ |
||
| 914 | "train_fn_to_labels['ID_079945c27']" |
||
| 915 | ] |
||
| 916 | }, |
||
| 917 | { |
||
| 918 | "cell_type": "code", |
||
| 919 | "execution_count": 16, |
||
| 920 | "metadata": {}, |
||
| 921 | "outputs": [], |
||
| 922 | "source": [ |
||
| 923 | "pickle.dump(train_fn_to_labels, open(f\"data/{stage}_train_fn_to_labels.pickle\", 'wb'))" |
||
| 924 | ] |
||
| 925 | }, |
||
| 926 | { |
||
| 927 | "cell_type": "markdown", |
||
| 928 | "metadata": {}, |
||
| 929 | "source": [ |
||
| 930 | "# Test dataset" |
||
| 931 | ] |
||
| 932 | }, |
||
| 933 | { |
||
| 934 | "cell_type": "code", |
||
| 935 | "execution_count": 17, |
||
| 936 | "metadata": {}, |
||
| 937 | "outputs": [], |
||
| 938 | "source": [ |
||
| 939 | "df = pd.read_csv(f\"data/{stage}_test_dicom.csv\")" |
||
| 940 | ] |
||
| 941 | }, |
||
| 942 | { |
||
| 943 | "cell_type": "code", |
||
| 944 | "execution_count": 18, |
||
| 945 | "metadata": {}, |
||
| 946 | "outputs": [ |
||
| 947 | { |
||
| 948 | "data": { |
||
| 949 | "text/html": [ |
||
| 950 | "<div>\n", |
||
| 951 | "<style scoped>\n", |
||
| 952 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 953 | " vertical-align: middle;\n", |
||
| 954 | " }\n", |
||
| 955 | "\n", |
||
| 956 | " .dataframe tbody tr th {\n", |
||
| 957 | " vertical-align: top;\n", |
||
| 958 | " }\n", |
||
| 959 | "\n", |
||
| 960 | " .dataframe thead th {\n", |
||
| 961 | " text-align: right;\n", |
||
| 962 | " }\n", |
||
| 963 | "</style>\n", |
||
| 964 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 965 | " <thead>\n", |
||
| 966 | " <tr style=\"text-align: right;\">\n", |
||
| 967 | " <th></th>\n", |
||
| 968 | " <th>Unnamed: 0</th>\n", |
||
| 969 | " <th>BitsAllocated</th>\n", |
||
| 970 | " <th>BitsStored</th>\n", |
||
| 971 | " <th>Columns</th>\n", |
||
| 972 | " <th>HighBit</th>\n", |
||
| 973 | " <th>ImageOrientationPatient_0</th>\n", |
||
| 974 | " <th>ImageOrientationPatient_1</th>\n", |
||
| 975 | " <th>ImageOrientationPatient_2</th>\n", |
||
| 976 | " <th>ImageOrientationPatient_3</th>\n", |
||
| 977 | " <th>ImageOrientationPatient_4</th>\n", |
||
| 978 | " <th>...</th>\n", |
||
| 979 | " <th>SamplesPerPixel</th>\n", |
||
| 980 | " <th>SeriesInstanceUID</th>\n", |
||
| 981 | " <th>StudyID</th>\n", |
||
| 982 | " <th>StudyInstanceUID</th>\n", |
||
| 983 | " <th>WindowCenter</th>\n", |
||
| 984 | " <th>WindowCenter_0</th>\n", |
||
| 985 | " <th>WindowCenter_1</th>\n", |
||
| 986 | " <th>WindowWidth</th>\n", |
||
| 987 | " <th>WindowWidth_0</th>\n", |
||
| 988 | " <th>WindowWidth_1</th>\n", |
||
| 989 | " </tr>\n", |
||
| 990 | " </thead>\n", |
||
| 991 | " <tbody>\n", |
||
| 992 | " <tr>\n", |
||
| 993 | " <th>0</th>\n", |
||
| 994 | " <td>0</td>\n", |
||
| 995 | " <td>16</td>\n", |
||
| 996 | " <td>12</td>\n", |
||
| 997 | " <td>512</td>\n", |
||
| 998 | " <td>11</td>\n", |
||
| 999 | " <td>1.0</td>\n", |
||
| 1000 | " <td>0.0</td>\n", |
||
| 1001 | " <td>0.0</td>\n", |
||
| 1002 | " <td>0.0</td>\n", |
||
| 1003 | " <td>0.981627</td>\n", |
||
| 1004 | " <td>...</td>\n", |
||
| 1005 | " <td>1</td>\n", |
||
| 1006 | " <td>ID_4d28912ba6</td>\n", |
||
| 1007 | " <td>NaN</td>\n", |
||
| 1008 | " <td>ID_1f6d1e8aeb</td>\n", |
||
| 1009 | " <td>NaN</td>\n", |
||
| 1010 | " <td>40.0</td>\n", |
||
| 1011 | " <td>40.0</td>\n", |
||
| 1012 | " <td>NaN</td>\n", |
||
| 1013 | " <td>80.0</td>\n", |
||
| 1014 | " <td>80.0</td>\n", |
||
| 1015 | " </tr>\n", |
||
| 1016 | " <tr>\n", |
||
| 1017 | " <th>1</th>\n", |
||
| 1018 | " <td>1</td>\n", |
||
| 1019 | " <td>16</td>\n", |
||
| 1020 | " <td>16</td>\n", |
||
| 1021 | " <td>512</td>\n", |
||
| 1022 | " <td>15</td>\n", |
||
| 1023 | " <td>1.0</td>\n", |
||
| 1024 | " <td>0.0</td>\n", |
||
| 1025 | " <td>0.0</td>\n", |
||
| 1026 | " <td>0.0</td>\n", |
||
| 1027 | " <td>0.987688</td>\n", |
||
| 1028 | " <td>...</td>\n", |
||
| 1029 | " <td>1</td>\n", |
||
| 1030 | " <td>ID_acabdeee86</td>\n", |
||
| 1031 | " <td>NaN</td>\n", |
||
| 1032 | " <td>ID_4a8d7ec19f</td>\n", |
||
| 1033 | " <td>30.0</td>\n", |
||
| 1034 | " <td>NaN</td>\n", |
||
| 1035 | " <td>NaN</td>\n", |
||
| 1036 | " <td>80.0</td>\n", |
||
| 1037 | " <td>NaN</td>\n", |
||
| 1038 | " <td>NaN</td>\n", |
||
| 1039 | " </tr>\n", |
||
| 1040 | " <tr>\n", |
||
| 1041 | " <th>2</th>\n", |
||
| 1042 | " <td>2</td>\n", |
||
| 1043 | " <td>16</td>\n", |
||
| 1044 | " <td>16</td>\n", |
||
| 1045 | " <td>512</td>\n", |
||
| 1046 | " <td>15</td>\n", |
||
| 1047 | " <td>1.0</td>\n", |
||
| 1048 | " <td>0.0</td>\n", |
||
| 1049 | " <td>0.0</td>\n", |
||
| 1050 | " <td>0.0</td>\n", |
||
| 1051 | " <td>0.927184</td>\n", |
||
| 1052 | " <td>...</td>\n", |
||
| 1053 | " <td>1</td>\n", |
||
| 1054 | " <td>ID_d00cee7f0c</td>\n", |
||
| 1055 | " <td>NaN</td>\n", |
||
| 1056 | " <td>ID_a6ca244172</td>\n", |
||
| 1057 | " <td>30.0</td>\n", |
||
| 1058 | " <td>NaN</td>\n", |
||
| 1059 | " <td>NaN</td>\n", |
||
| 1060 | " <td>80.0</td>\n", |
||
| 1061 | " <td>NaN</td>\n", |
||
| 1062 | " <td>NaN</td>\n", |
||
| 1063 | " </tr>\n", |
||
| 1064 | " <tr>\n", |
||
| 1065 | " <th>3</th>\n", |
||
| 1066 | " <td>3</td>\n", |
||
| 1067 | " <td>16</td>\n", |
||
| 1068 | " <td>16</td>\n", |
||
| 1069 | " <td>512</td>\n", |
||
| 1070 | " <td>15</td>\n", |
||
| 1071 | " <td>1.0</td>\n", |
||
| 1072 | " <td>0.0</td>\n", |
||
| 1073 | " <td>0.0</td>\n", |
||
| 1074 | " <td>0.0</td>\n", |
||
| 1075 | " <td>0.986286</td>\n", |
||
| 1076 | " <td>...</td>\n", |
||
| 1077 | " <td>1</td>\n", |
||
| 1078 | " <td>ID_a52a0112d5</td>\n", |
||
| 1079 | " <td>NaN</td>\n", |
||
| 1080 | " <td>ID_fa950a03af</td>\n", |
||
| 1081 | " <td>30.0</td>\n", |
||
| 1082 | " <td>NaN</td>\n", |
||
| 1083 | " <td>NaN</td>\n", |
||
| 1084 | " <td>80.0</td>\n", |
||
| 1085 | " <td>NaN</td>\n", |
||
| 1086 | " <td>NaN</td>\n", |
||
| 1087 | " </tr>\n", |
||
| 1088 | " <tr>\n", |
||
| 1089 | " <th>4</th>\n", |
||
| 1090 | " <td>4</td>\n", |
||
| 1091 | " <td>16</td>\n", |
||
| 1092 | " <td>12</td>\n", |
||
| 1093 | " <td>512</td>\n", |
||
| 1094 | " <td>11</td>\n", |
||
| 1095 | " <td>1.0</td>\n", |
||
| 1096 | " <td>0.0</td>\n", |
||
| 1097 | " <td>0.0</td>\n", |
||
| 1098 | " <td>0.0</td>\n", |
||
| 1099 | " <td>1.000000</td>\n", |
||
| 1100 | " <td>...</td>\n", |
||
| 1101 | " <td>1</td>\n", |
||
| 1102 | " <td>ID_f552d3b922</td>\n", |
||
| 1103 | " <td>NaN</td>\n", |
||
| 1104 | " <td>ID_965d8b3d8e</td>\n", |
||
| 1105 | " <td>NaN</td>\n", |
||
| 1106 | " <td>36.0</td>\n", |
||
| 1107 | " <td>36.0</td>\n", |
||
| 1108 | " <td>NaN</td>\n", |
||
| 1109 | " <td>80.0</td>\n", |
||
| 1110 | " <td>80.0</td>\n", |
||
| 1111 | " </tr>\n", |
||
| 1112 | " </tbody>\n", |
||
| 1113 | "</table>\n", |
||
| 1114 | "<p>5 rows × 34 columns</p>\n", |
||
| 1115 | "</div>" |
||
| 1116 | ], |
||
| 1117 | "text/plain": [ |
||
| 1118 | " Unnamed: 0 BitsAllocated BitsStored Columns HighBit \\\n", |
||
| 1119 | "0 0 16 12 512 11 \n", |
||
| 1120 | "1 1 16 16 512 15 \n", |
||
| 1121 | "2 2 16 16 512 15 \n", |
||
| 1122 | "3 3 16 16 512 15 \n", |
||
| 1123 | "4 4 16 12 512 11 \n", |
||
| 1124 | "\n", |
||
| 1125 | " ImageOrientationPatient_0 ImageOrientationPatient_1 \\\n", |
||
| 1126 | "0 1.0 0.0 \n", |
||
| 1127 | "1 1.0 0.0 \n", |
||
| 1128 | "2 1.0 0.0 \n", |
||
| 1129 | "3 1.0 0.0 \n", |
||
| 1130 | "4 1.0 0.0 \n", |
||
| 1131 | "\n", |
||
| 1132 | " ImageOrientationPatient_2 ImageOrientationPatient_3 \\\n", |
||
| 1133 | "0 0.0 0.0 \n", |
||
| 1134 | "1 0.0 0.0 \n", |
||
| 1135 | "2 0.0 0.0 \n", |
||
| 1136 | "3 0.0 0.0 \n", |
||
| 1137 | "4 0.0 0.0 \n", |
||
| 1138 | "\n", |
||
| 1139 | " ImageOrientationPatient_4 ... SamplesPerPixel SeriesInstanceUID \\\n", |
||
| 1140 | "0 0.981627 ... 1 ID_4d28912ba6 \n", |
||
| 1141 | "1 0.987688 ... 1 ID_acabdeee86 \n", |
||
| 1142 | "2 0.927184 ... 1 ID_d00cee7f0c \n", |
||
| 1143 | "3 0.986286 ... 1 ID_a52a0112d5 \n", |
||
| 1144 | "4 1.000000 ... 1 ID_f552d3b922 \n", |
||
| 1145 | "\n", |
||
| 1146 | " StudyID StudyInstanceUID WindowCenter WindowCenter_0 WindowCenter_1 \\\n", |
||
| 1147 | "0 NaN ID_1f6d1e8aeb NaN 40.0 40.0 \n", |
||
| 1148 | "1 NaN ID_4a8d7ec19f 30.0 NaN NaN \n", |
||
| 1149 | "2 NaN ID_a6ca244172 30.0 NaN NaN \n", |
||
| 1150 | "3 NaN ID_fa950a03af 30.0 NaN NaN \n", |
||
| 1151 | "4 NaN ID_965d8b3d8e NaN 36.0 36.0 \n", |
||
| 1152 | "\n", |
||
| 1153 | " WindowWidth WindowWidth_0 WindowWidth_1 \n", |
||
| 1154 | "0 NaN 80.0 80.0 \n", |
||
| 1155 | "1 80.0 NaN NaN \n", |
||
| 1156 | "2 80.0 NaN NaN \n", |
||
| 1157 | "3 80.0 NaN NaN \n", |
||
| 1158 | "4 NaN 80.0 80.0 \n", |
||
| 1159 | "\n", |
||
| 1160 | "[5 rows x 34 columns]" |
||
| 1161 | ] |
||
| 1162 | }, |
||
| 1163 | "execution_count": 18, |
||
| 1164 | "metadata": {}, |
||
| 1165 | "output_type": "execute_result" |
||
| 1166 | } |
||
| 1167 | ], |
||
| 1168 | "source": [ |
||
| 1169 | "df.head()" |
||
| 1170 | ] |
||
| 1171 | }, |
||
| 1172 | { |
||
| 1173 | "cell_type": "code", |
||
| 1174 | "execution_count": 19, |
||
| 1175 | "metadata": {}, |
||
| 1176 | "outputs": [ |
||
| 1177 | { |
||
| 1178 | "data": { |
||
| 1179 | "text/plain": [ |
||
| 1180 | "3518" |
||
| 1181 | ] |
||
| 1182 | }, |
||
| 1183 | "execution_count": 19, |
||
| 1184 | "metadata": {}, |
||
| 1185 | "output_type": "execute_result" |
||
| 1186 | } |
||
| 1187 | ], |
||
| 1188 | "source": [ |
||
| 1189 | "# sort, then group by (order is preserver within groups)\n", |
||
| 1190 | "gs = df.sort_values('ImagePositionPatient_2').groupby('SeriesInstanceUID')\n", |
||
| 1191 | "len(gs)" |
||
| 1192 | ] |
||
| 1193 | }, |
||
| 1194 | { |
||
| 1195 | "cell_type": "code", |
||
| 1196 | "execution_count": 20, |
||
| 1197 | "metadata": {}, |
||
| 1198 | "outputs": [], |
||
| 1199 | "source": [ |
||
| 1200 | "def rename_test_group(subg):\n", |
||
| 1201 | " ix = 0\n", |
||
| 1202 | " total = len(subg)\n", |
||
| 1203 | " for index, row in subg.iterrows():\n", |
||
| 1204 | " cur_fn = row['SOPInstanceUID']\n", |
||
| 1205 | " new_fn = f\"{row['SeriesInstanceUID']}_{ix:03}_{total:03}_{cur_fn}\"\n", |
||
| 1206 | " ix += 1\n", |
||
| 1207 | " Path(f'data/unzip_renamed/{stage}_test_images/{cur_fn}.dcm').rename(f'data/unzip_renamed/{stage}_test_images/{new_fn}.dcm')\n" |
||
| 1208 | ] |
||
| 1209 | }, |
||
| 1210 | { |
||
| 1211 | "cell_type": "code", |
||
| 1212 | "execution_count": 21, |
||
| 1213 | "metadata": {}, |
||
| 1214 | "outputs": [], |
||
| 1215 | "source": [ |
||
| 1216 | "test_study_ix_to_fn = defaultdict(list)\n", |
||
| 1217 | "test_fn_to_study_ix = {}\n", |
||
| 1218 | "\n", |
||
| 1219 | "for name, subg in gs:\n", |
||
| 1220 | " #rename_test_group(subg)\n", |
||
| 1221 | " index_group(subg, test_study_ix_to_fn, test_fn_to_study_ix)\n" |
||
| 1222 | ] |
||
| 1223 | }, |
||
| 1224 | { |
||
| 1225 | "cell_type": "code", |
||
| 1226 | "execution_count": 22, |
||
| 1227 | "metadata": {}, |
||
| 1228 | "outputs": [], |
||
| 1229 | "source": [ |
||
| 1230 | "pickle.dump(test_study_ix_to_fn, open(f\"data/{stage}_test_study_ix_to_fn.pickle\", \"wb\" ))\n", |
||
| 1231 | "pickle.dump(test_fn_to_study_ix, open(f\"data/{stage}_test_fn_to_study_ix.pickle\", \"wb\" ))" |
||
| 1232 | ] |
||
| 1233 | }, |
||
| 1234 | { |
||
| 1235 | "cell_type": "code", |
||
| 1236 | "execution_count": 23, |
||
| 1237 | "metadata": {}, |
||
| 1238 | "outputs": [], |
||
| 1239 | "source": [ |
||
| 1240 | "study_ix_to_fn = { **train_study_ix_to_fn, **test_study_ix_to_fn }\n", |
||
| 1241 | "fn_to_study_ix = { **train_fn_to_study_ix, **test_fn_to_study_ix }" |
||
| 1242 | ] |
||
| 1243 | }, |
||
| 1244 | { |
||
| 1245 | "cell_type": "code", |
||
| 1246 | "execution_count": 24, |
||
| 1247 | "metadata": {}, |
||
| 1248 | "outputs": [], |
||
| 1249 | "source": [ |
||
| 1250 | "pickle.dump(study_ix_to_fn, open(f\"data/{stage}_study_ix_to_fn.pickle\", \"wb\" ))\n", |
||
| 1251 | "pickle.dump(fn_to_study_ix, open(f\"data/{stage}_fn_to_study_ix.pickle\", \"wb\" ))" |
||
| 1252 | ] |
||
| 1253 | } |
||
| 1254 | ], |
||
| 1255 | "metadata": { |
||
| 1256 | "kernelspec": { |
||
| 1257 | "display_name": "Python 3", |
||
| 1258 | "language": "python", |
||
| 1259 | "name": "python3" |
||
| 1260 | }, |
||
| 1261 | "language_info": { |
||
| 1262 | "codemirror_mode": { |
||
| 1263 | "name": "ipython", |
||
| 1264 | "version": 3 |
||
| 1265 | }, |
||
| 1266 | "file_extension": ".py", |
||
| 1267 | "mimetype": "text/x-python", |
||
| 1268 | "name": "python", |
||
| 1269 | "nbconvert_exporter": "python", |
||
| 1270 | "pygments_lexer": "ipython3", |
||
| 1271 | "version": "3.7.3" |
||
| 1272 | } |
||
| 1273 | }, |
||
| 1274 | "nbformat": 4, |
||
| 1275 | "nbformat_minor": 2 |
||
| 1276 | } |