├── MRI_ImageClassification_Keras.ipynb
├── MRI_ImageClassification_Keras_Core.ipynb
└── README.md


/MRI_ImageClassification_Keras_Core.ipynb:
--------------------------------------------------------------------------------
  1 | {
  2 |   "nbformat": 4,
  3 |   "nbformat_minor": 0,
  4 |   "metadata": {
  5 |     "colab": {
  6 |       "provenance": [],
  7 |       "gpuType": "T4"
  8 |     },
  9 |     "kernelspec": {
 10 |       "name": "python3",
 11 |       "display_name": "Python 3"
 12 |     },
 13 |     "language_info": {
 14 |       "name": "python"
 15 |     },
 16 |     "accelerator": "GPU"
 17 |   },
 18 |   "cells": [
 19 |     {
 20 |       "cell_type": "markdown",
 21 |       "source": [
 22 |         "## Installing the required libraries"
 23 |       ],
 24 |       "metadata": {
 25 |         "id": "IG2ammtIxmd0"
 26 |       }
 27 |     },
 28 |     {
 29 |       "cell_type": "code",
 30 |       "source": [
 31 |         "!pip install --quiet nibabel simpleitk nilearn transformers vit-keras tensorflow-addons keras-core"
 32 |       ],
 33 |       "metadata": {
 34 |         "id": "xCULO5hVgUaS",
 35 |         "colab": {
 36 |           "base_uri": "https://localhost:8080/"
 37 |         },
 38 |         "outputId": "5c1e2512-0cb8-43c0-ba48-0fa8cc477525"
 39 |       },
 40 |       "execution_count": null,
 41 |       "outputs": [
 42 |         {
 43 |           "output_type": "stream",
 44 |           "name": "stdout",
 45 |           "text": [
 46 |             "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m52.7/52.7 MB\u001b[0m \u001b[31m15.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
 47 |             "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m10.3/10.3 MB\u001b[0m \u001b[31m110.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
 48 |             "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.1/7.1 MB\u001b[0m \u001b[31m118.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
 49 |             "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m591.0/591.0 kB\u001b[0m \u001b[31m49.3 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
 50 |             "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m3.3/3.3 MB\u001b[0m \u001b[31m112.5 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
 51 |             "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m224.5/224.5 kB\u001b[0m \u001b[31m17.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
 52 |             "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m7.8/7.8 MB\u001b[0m \u001b[31m128.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
 53 |             "\u001b[?25h  Preparing metadata (setup.py) ... \u001b[?25l\u001b[?25hdone\n",
 54 |             "  Building wheel for validators (setup.py) ... \u001b[?25l\u001b[?25hdone\n"
 55 |           ]
 56 |         }
 57 |       ]
 58 |     },
 59 |     {
 60 |       "cell_type": "markdown",
 61 |       "source": [
 62 |         "## Downloading and processing the abide dataset using the nilearn library"
 63 |       ],
 64 |       "metadata": {
 65 |         "id": "C3g0X6SSxuDB"
 66 |       }
 67 |     },
 68 |     {
 69 |       "cell_type": "code",
 70 |       "source": [
 71 |         "#!/usr/bin/env python\n",
 72 |         "\n",
 73 |         "from nilearn import datasets\n",
 74 |         "from nilearn.input_data import NiftiLabelsMasker\n",
 75 |         "from nilearn.connectome import ConnectivityMeasure\n",
 76 |         "from argparse import ArgumentParser\n",
 77 |         "import numpy as np\n",
 78 |         "from sklearn.decomposition import PCA\n",
 79 |         "import os\n",
 80 |         "import pandas as pd\n",
 81 |         "\n",
 82 |         "def prepare_data(data_dir, output_dir, pipeline = \"cpac\", quality_checked = True):\n",
 83 |         "    # get dataset\n",
 84 |         "    print(\"Loading dataset...\")\n",
 85 |         "    abide = datasets.fetch_abide_pcp(data_dir = data_dir,\n",
 86 |         "                                     pipeline = pipeline,\n",
 87 |         "                                     n_subjects = 1,\n",
 88 |         "                                     quality_checked = quality_checked)\n",
 89 |         "    # make list of filenames\n",
 90 |         "    fmri_filenames = abide.func_preproc\n",
 91 |         "\n",
 92 |         "    # load atlas\n",
 93 |         "    multiscale = datasets.fetch_atlas_basc_multiscale_2015()\n",
 94 |         "    atlas_filename = multiscale.scale064\n",
 95 |         "\n",
 96 |         "    # initialize masker object\n",
 97 |         "    masker = NiftiLabelsMasker(labels_img=atlas_filename,\n",
 98 |         "                               standardize=True,\n",
 99 |         "                               memory='nilearn_cache',\n",
100 |         "                               verbose=0)\n",
101 |         "\n",
102 |         "    # initialize correlation measure\n",
103 |         "    correlation_measure = ConnectivityMeasure(kind='correlation', vectorize=True,\n",
104 |         "                                             discard_diagonal=True)\n",
105 |         "\n",
106 |         "    try: # check if feature file already exists\n",
107 |         "        # load features\n",
108 |         "        feat_file = os.path.join(output_dir, 'ABIDE_BASC064_features.npz')\n",
109 |         "        X_features = np.load(feat_file)['a']\n",
110 |         "        print(\"Feature file found.\")\n",
111 |         "\n",
112 |         "    except: # if not, extract features\n",
113 |         "        X_features = [] # To contain upper half of matrix as 1d array\n",
114 |         "        print(\"No feature file found. Extracting features...\")\n",
115 |         "\n",
116 |         "        for i,sub in enumerate(fmri_filenames):\n",
117 |         "            # extract the timeseries from the ROIs in the atlas\n",
118 |         "            time_series = masker.fit_transform(sub)\n",
119 |         "            # create a region x region correlation matrix\n",
120 |         "            correlation_matrix = correlation_measure.fit_transform([time_series])[0]\n",
121 |         "            # add to our container\n",
122 |         "            X_features.append(correlation_matrix)\n",
123 |         "            # keep track of status\n",
124 |         "            print('finished extracting %s of %s'%(i+1,len(fmri_filenames)))\n",
125 |         "        # Save features\n",
126 |         "        np.savez_compressed(os.path.join(output_dir, 'ABIDE_BASC064_features'),\n",
127 |         "                                         a = X_features)\n",
128 |         "\n",
129 |         "    # Dimensionality reduction of features with PCA\n",
130 |         "    # print(\"Running PCA...\")\n",
131 |         "    # pca = PCA(0.99).fit(X_features) # keeping 99% of variance\n",
132 |         "    # X_features_pca = pca.transform(X_features)\n",
133 |         "\n",
134 |         "    # Transform phenotypic data into dataframe\n",
135 |         "    abide_pheno = pd.DataFrame(abide.phenotypic)\n",
136 |         "\n",
137 |         "    # Get the target vector\n",
138 |         "    y_target = abide_pheno['DX_GROUP']\n",
139 |         "\n",
140 |         "    return(X_features, y_target)\n",
141 |         "\n"
142 |       ],
143 |       "metadata": {
144 |         "id": "1cy5dnOqWAIT",
145 |         "colab": {
146 |           "base_uri": "https://localhost:8080/"
147 |         },
148 |         "outputId": "a111eb8f-ca93-4a92-a10f-429bece03db2"
149 |       },
150 |       "execution_count": null,
151 |       "outputs": [
152 |         {
153 |           "output_type": "stream",
154 |           "name": "stderr",
155 |           "text": [
156 |             "/usr/local/lib/python3.10/dist-packages/nilearn/input_data/__init__.py:23: FutureWarning: The import path 'nilearn.input_data' is deprecated in version 0.9. Importing from 'nilearn.input_data' will be possible at least until release 0.13.0. Please import from 'nilearn.maskers' instead.\n",
157 |             "  warnings.warn(message, FutureWarning)\n"
158 |           ]
159 |         }
160 |       ]
161 |     },
162 |     {
163 |       "cell_type": "code",
164 |       "source": [
165 |         "#Import modules\n",
166 |         "import numpy as np\n",
167 |         "from sklearn.model_selection import GroupKFold\n",
168 |         "import os\n",
169 |         "\n",
170 |         "#Define data and output directories\n",
171 |         "data_dir = 'data/'\n",
172 |         "output_dir = data_dir\n",
173 |         "\n",
174 |         "X, y = prepare_data(data_dir,output_dir)\n",
175 |         "\n",
176 |         "# logo = GroupKFold(n_splits=10)\n",
177 |         "# logo.get_n_splits(X, y, groups)"
178 |       ],
179 |       "metadata": {
180 |         "id": "is7VuFgKVfRA"
181 |       },
182 |       "execution_count": null,
183 |       "outputs": []
184 |     },
185 |     {
186 |       "cell_type": "code",
187 |       "source": [
188 |         "np.save('x.npy',X)\n",
189 |         "np.save('y.npy',y)"
190 |       ],
191 |       "metadata": {
192 |         "id": "PcZoaXXs4_bb"
193 |       },
194 |       "execution_count": null,
195 |       "outputs": []
196 |     },
197 |     {
198 |       "cell_type": "code",
199 |       "source": [
200 |         "import numpy as np\n",
201 |         "X = np.load('x.npy')\n",
202 |         "y=np.load('y.npy')"
203 |       ],
204 |       "metadata": {
205 |         "id": "bszZFPsRi6Xl"
206 |       },
207 |       "execution_count": null,
208 |       "outputs": []
209 |     },
210 |     {
211 |       "cell_type": "code",
212 |       "source": [
213 |         "X= np.array(X)\n",
214 |         "y = np.array(y)"
215 |       ],
216 |       "metadata": {
217 |         "id": "607wWB46g9TR"
218 |       },
219 |       "execution_count": null,
220 |       "outputs": []
221 |     },
222 |     {
223 |       "cell_type": "code",
224 |       "source": [
225 |         "X.shape,y.shape"
226 |       ],
227 |       "metadata": {
228 |         "colab": {
229 |           "base_uri": "https://localhost:8080/"
230 |         },
231 |         "id": "3SkOA2mIgxNO",
232 |         "outputId": "b73964da-eae5-4c07-be41-73d3908e6798"
233 |       },
234 |       "execution_count": null,
235 |       "outputs": [
236 |         {
237 |           "output_type": "execute_result",
238 |           "data": {
239 |             "text/plain": [
240 |               "((400, 2016), (400,))"
241 |             ]
242 |           },
243 |           "metadata": {},
244 |           "execution_count": 9
245 |         }
246 |       ]
247 |     },
248 |     {
249 |       "cell_type": "code",
250 |       "source": [
251 |         "X = np.reshape(X,(X.shape[0],42,-1))\n",
252 |         "y=y-1"
253 |       ],
254 |       "metadata": {
255 |         "id": "QtqHoZ5F7NFb"
256 |       },
257 |       "execution_count": null,
258 |       "outputs": []
259 |     },
260 |     {
261 |       "cell_type": "code",
262 |       "source": [
263 |         "b = np.zeros((y.size, int(y.max()) + 1))\n",
264 |         "b[np.arange(y.size), y.astype(np.uint)] = 1\n",
265 |         "b.shape"
266 |       ],
267 |       "metadata": {
268 |         "colab": {
269 |           "base_uri": "https://localhost:8080/"
270 |         },
271 |         "id": "RfgXm_RbLkxg",
272 |         "outputId": "f5db20f3-729c-4ac0-d19a-e20eb991e10e"
273 |       },
274 |       "execution_count": null,
275 |       "outputs": [
276 |         {
277 |           "output_type": "execute_result",
278 |           "data": {
279 |             "text/plain": [
280 |               "(400, 2)"
281 |             ]
282 |           },
283 |           "metadata": {},
284 |           "execution_count": 11
285 |         }
286 |       ]
287 |     },
288 |     {
289 |       "cell_type": "code",
290 |       "source": [
291 |         "y=b\n",
292 |         "y.shape"
293 |       ],
294 |       "metadata": {
295 |         "colab": {
296 |           "base_uri": "https://localhost:8080/"
297 |         },
298 |         "id": "ZgugMzuqLmol",
299 |         "outputId": "d6c21db5-2de1-46d6-a945-f6e74a630d1d"
300 |       },
301 |       "execution_count": null,
302 |       "outputs": [
303 |         {
304 |           "output_type": "execute_result",
305 |           "data": {
306 |             "text/plain": [
307 |               "(400, 2)"
308 |             ]
309 |           },
310 |           "metadata": {},
311 |           "execution_count": 12
312 |         }
313 |       ]
314 |     },
315 |     {
316 |       "cell_type": "markdown",
317 |       "source": [
318 |         "Making the dataset ready to be used for the model training"
319 |       ],
320 |       "metadata": {
321 |         "id": "7fSCaZtXx-Ba"
322 |       }
323 |     },
324 |     {
325 |       "cell_type": "code",
326 |       "source": [
327 |         "Xx = np.stack([X,X,X],axis=3)\n",
328 |         "Xx = np.resize(Xx,(Xx.shape[0],224,224,3))\n",
329 |         "Xx.shape"
330 |       ],
331 |       "metadata": {
332 |         "colab": {
333 |           "base_uri": "https://localhost:8080/"
334 |         },
335 |         "id": "Iq8R7yPv_hoe",
336 |         "outputId": "900ddbb8-80cd-48e3-b0a4-185c8a477915"
337 |       },
338 |       "execution_count": null,
339 |       "outputs": [
340 |         {
341 |           "output_type": "execute_result",
342 |           "data": {
343 |             "text/plain": [
344 |               "(400, 224, 224, 3)"
345 |             ]
346 |           },
347 |           "metadata": {},
348 |           "execution_count": 13
349 |         }
350 |       ]
351 |     },
352 |     {
353 |       "cell_type": "code",
354 |       "source": [
355 |         "from google.colab.patches import cv2_imshow\n",
356 |         "cv2_imshow(Xx[10]*255)"
357 |       ],
358 |       "metadata": {
359 |         "colab": {
360 |           "base_uri": "https://localhost:8080/",
361 |           "height": 241
362 |         },
363 |         "id": "3qeec3FaOiRb",
364 |         "outputId": "f990a4b1-007c-4f29-82c4-abcc88206b47"
365 |       },
366 |       "execution_count": null,
367 |       "outputs": [
368 |         {
369 |           "output_type": "display_data",
370 |           "data": {
371 |             "text/plain": [
372 |               "<PIL.Image.Image image mode=RGB size=224x224 at 0x7FD20969BEB0>"
373 |             ],
374 |             "image/png": "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\n"
375 |           },
376 |           "metadata": {}
377 |         }
378 |       ]
379 |     },
380 |     {
381 |       "cell_type": "code",
382 |       "source": [
383 |         "import cv2\n",
384 |         "cv2.imwrite('test.jpg',Xx[0])"
385 |       ],
386 |       "metadata": {
387 |         "colab": {
388 |           "base_uri": "https://localhost:8080/"
389 |         },
390 |         "id": "v6tN9pEqOt3v",
391 |         "outputId": "cfccbe24-5d92-40bc-d748-5d04b62c71f2"
392 |       },
393 |       "execution_count": null,
394 |       "outputs": [
395 |         {
396 |           "output_type": "execute_result",
397 |           "data": {
398 |             "text/plain": [
399 |               "True"
400 |             ]
401 |           },
402 |           "metadata": {},
403 |           "execution_count": 18
404 |         }
405 |       ]
406 |     },
407 |     {
408 |       "cell_type": "code",
409 |       "source": [
410 |         "np.save(Xx[0].shape)"
411 |       ],
412 |       "metadata": {
413 |         "colab": {
414 |           "base_uri": "https://localhost:8080/"
415 |         },
416 |         "id": "df2r4BCROaCi",
417 |         "outputId": "7045262d-0fc0-4ff8-cf89-76c005cc7815"
418 |       },
419 |       "execution_count": null,
420 |       "outputs": [
421 |         {
422 |           "output_type": "execute_result",
423 |           "data": {
424 |             "text/plain": [
425 |               "(224, 224, 3)"
426 |             ]
427 |           },
428 |           "metadata": {},
429 |           "execution_count": 14
430 |         }
431 |       ]
432 |     },
433 |     {
434 |       "cell_type": "markdown",
435 |       "source": [
436 |         "## Building a teacher student model structure for the pretraining of the ViT model"
437 |       ],
438 |       "metadata": {
439 |         "id": "_2QqNKNxyfOM"
440 |       }
441 |     },
442 |     {
443 |       "cell_type": "code",
444 |       "source": [
445 |         "import tensorflow as tf\n",
446 |         "from keras_core import Model, Sequential\n",
447 |         "from keras_core.layers import Layer\n",
448 |         "import keras_core.layers as nn\n",
449 |         "import keras_core as keras"
450 |       ],
451 |       "metadata": {
452 |         "id": "Kl0YNY2Ez2nh"
453 |       },
454 |       "execution_count": null,
455 |       "outputs": []
456 |     },
457 |     {
458 |       "cell_type": "code",
459 |       "source": [
460 |         "from vit_keras import vit\n",
461 |         "vit_model = vit.vit_b32(\n",
462 |         "        image_size = (224,224),\n",
463 |         "        pretrained = True,\n",
464 |         "        include_top = True)"
465 |       ],
466 |       "metadata": {
467 |         "id": "_h8Rwp226q1R",
468 |         "colab": {
469 |           "base_uri": "https://localhost:8080/"
470 |         },
471 |         "outputId": "5112fca9-bf2a-4564-b8cc-ee33b82fe2b2"
472 |       },
473 |       "execution_count": null,
474 |       "outputs": [
475 |         {
476 |           "output_type": "stream",
477 |           "name": "stderr",
478 |           "text": [
479 |             "/usr/local/lib/python3.10/dist-packages/tensorflow_addons/utils/tfa_eol_msg.py:23: UserWarning: \n",
480 |             "\n",
481 |             "TensorFlow Addons (TFA) has ended development and introduction of new features.\n",
482 |             "TFA has entered a minimal maintenance and release mode until a planned end of life in May 2024.\n",
483 |             "Please modify downstream libraries to take dependencies from other repositories in our TensorFlow community (e.g. Keras, Keras-CV, and Keras-NLP). \n",
484 |             "\n",
485 |             "For more information see: https://github.com/tensorflow/addons/issues/2807 \n",
486 |             "\n",
487 |             "  warnings.warn(\n"
488 |           ]
489 |         },
490 |         {
491 |           "output_type": "stream",
492 |           "name": "stdout",
493 |           "text": [
494 |             "Downloading data from https://github.com/faustomorales/vit-keras/releases/download/dl/ViT-B_32_imagenet21k+imagenet2012.npz\n",
495 |             "353253686/353253686 [==============================] - 3s 0us/step\n"
496 |           ]
497 |         },
498 |         {
499 |           "output_type": "stream",
500 |           "name": "stderr",
501 |           "text": [
502 |             "/usr/local/lib/python3.10/dist-packages/vit_keras/utils.py:81: UserWarning: Resizing position embeddings from 12, 12 to 7, 7\n",
503 |             "  warnings.warn(\n"
504 |           ]
505 |         }
506 |       ]
507 |     },
508 |     {
509 |       "cell_type": "code",
510 |       "source": [
511 |         "import tensorflow as tf\n",
512 |         "base_cnn = keras.applications.resnet50.ResNet50(\n",
513 |         "    include_top=True,\n",
514 |         "    weights='imagenet',\n",
515 |         ")"
516 |       ],
517 |       "metadata": {
518 |         "id": "fCeIn1QnhZS_",
519 |         "colab": {
520 |           "base_uri": "https://localhost:8080/"
521 |         },
522 |         "outputId": "a083a706-32ed-4ac1-f83f-5b33fee63246"
523 |       },
524 |       "execution_count": null,
525 |       "outputs": [
526 |         {
527 |           "output_type": "stream",
528 |           "name": "stdout",
529 |           "text": [
530 |             "Downloading data from https://storage.googleapis.com/tensorflow/keras-applications/resnet/resnet50_weights_tf_dim_ordering_tf_kernels.h5\n",
531 |             "102967424/102967424 [==============================] - 1s 0us/step\n"
532 |           ]
533 |         }
534 |       ]
535 |     },
536 |     {
537 |       "cell_type": "code",
538 |       "source": [
539 |         "def loss_fn(x, y):\n",
540 |         "    x =  keras.utils.normalize(x, axis=-1, order=2)\n",
541 |         "    y =  keras.utils.normalize(y, axis=-1, order=2)\n",
542 |         "    return 2 - 2 * tf.reduce_sum((x * y))"
543 |       ],
544 |       "metadata": {
545 |         "id": "WQvjMGGex7mn"
546 |       },
547 |       "execution_count": null,
548 |       "outputs": []
549 |     },
550 |     {
551 |       "cell_type": "code",
552 |       "source": [
553 |         "optimizer1 = keras.optimizers.Adam()\n",
554 |         "optimizer2 = keras.optimizers.Adam()"
555 |       ],
556 |       "metadata": {
557 |         "id": "3FvXucTCzsNO"
558 |       },
559 |       "execution_count": null,
560 |       "outputs": []
561 |     },
562 |     {
563 |       "cell_type": "code",
564 |       "source": [
565 |         "import numpy as np\n",
566 |         "tf.config.run_functions_eagerly(True)\n",
567 |         "@tf.function\n",
568 |         "def train_step(x, y):\n",
569 |         "    with tf.GradientTape(persistent=True) as tape:\n",
570 |         "        logits_cnn = base_cnn(x)\n",
571 |         "        logits_vit = vit_model(x)\n",
572 |         "        loss_value = loss_fn(logits_cnn, logits_vit)\n",
573 |         "    grads1 = tape.gradient(loss_value, base_cnn.trainable_weights)\n",
574 |         "    optimizer1.apply_gradients(zip(grads1, base_cnn.trainable_weights))\n",
575 |         "    grads2 = tape.gradient(loss_value, vit_model.trainable_weights)\n",
576 |         "    optimizer2.apply_gradients(zip(grads2, vit_model.trainable_weights))\n",
577 |         "    return loss_value\n",
578 |         "\n",
579 |         "import time\n",
580 |         "train_loss_list=[]\n",
581 |         "epochs = 10\n",
582 |         "for epoch in range(epochs):\n",
583 |         "    print(\"\\nStart of epoch %d\" % (epoch,))\n",
584 |         "    start_time = time.time()\n",
585 |         "    train_loss = []\n",
586 |         "    x_batch_train=[]\n",
587 |         "    y_batch_train=[]\n",
588 |         "    for step, (t, z) in enumerate(zip(Xx,y)):\n",
589 |         "        x_batch_train.append(t)\n",
590 |         "        y_batch_train.append(z)\n",
591 |         "        if len(x_batch_train)==32:\n",
592 |         "          loss_value = train_step(np.array(x_batch_train), np.array(y_batch_train))\n",
593 |         "          train_loss.append(float(loss_value))\n",
594 |         "          x_batch_train=[]\n",
595 |         "          y_batch_train=[]\n",
596 |         "          if step % 100 == 0:\n",
597 |         "              print(\n",
598 |         "                  \"Training loss (for one batch) at step %d: %.4f\"\n",
599 |         "                  % (step, float(loss_value))\n",
600 |         "              )\n",
601 |         "\n"
602 |       ],
603 |       "metadata": {
604 |         "id": "auF8V0vmv5ca"
605 |       },
606 |       "execution_count": null,
607 |       "outputs": []
608 |     },
609 |     {
610 |       "cell_type": "code",
611 |       "source": [
612 |         "loss_value"
613 |       ],
614 |       "metadata": {
615 |         "colab": {
616 |           "base_uri": "https://localhost:8080/"
617 |         },
618 |         "id": "LTdnl_N4lBHm",
619 |         "outputId": "3d2441a5-0132-4eff-96e8-8322cbe1f822"
620 |       },
621 |       "execution_count": null,
622 |       "outputs": [
623 |         {
624 |           "output_type": "execute_result",
625 |           "data": {
626 |             "text/plain": [
627 |               "<tf.Tensor: shape=(), dtype=float32, numpy=-54.29072>"
628 |             ]
629 |           },
630 |           "metadata": {},
631 |           "execution_count": 15
632 |         }
633 |       ]
634 |     },
635 |     {
636 |       "cell_type": "code",
637 |       "source": [
638 |         "model_cnn_sup = Sequential([\n",
639 |         "    base_cnn,\n",
640 |         "    nn.Dense(2,activation='softmax')\n",
641 |         "])"
642 |       ],
643 |       "metadata": {
644 |         "id": "4lh3B3JYDgq4"
645 |       },
646 |       "execution_count": null,
647 |       "outputs": []
648 |     },
649 |     {
650 |       "cell_type": "code",
651 |       "source": [
652 |         "!pip install --quiet focal-loss"
653 |       ],
654 |       "metadata": {
655 |         "id": "RM3m8l-1M5Hh"
656 |       },
657 |       "execution_count": null,
658 |       "outputs": []
659 |     },
660 |     {
661 |       "cell_type": "markdown",
662 |       "source": [
663 |         "## Fine-tuning the cnn and vit models using the Binary Focal loss"
664 |       ],
665 |       "metadata": {
666 |         "id": "RCJJpjDIyouJ"
667 |       }
668 |     },
669 |     {
670 |       "cell_type": "code",
671 |       "source": [
672 |         "from focal_loss import BinaryFocalLoss"
673 |       ],
674 |       "metadata": {
675 |         "id": "b-YnOkUFM2nC"
676 |       },
677 |       "execution_count": null,
678 |       "outputs": []
679 |     },
680 |     {
681 |       "cell_type": "code",
682 |       "source": [
683 |         "model_cnn_sup.compile(loss=BinaryFocalLoss(gamma=2),optimizer='adam',metrics='accuracy')"
684 |       ],
685 |       "metadata": {
686 |         "id": "404AKtSgJxsA"
687 |       },
688 |       "execution_count": null,
689 |       "outputs": []
690 |     },
691 |     {
692 |       "cell_type": "code",
693 |       "source": [
694 |         "model_cnn_sup.fit(Xx,y,epochs=5,verbose=1,batch_size=32)"
695 |       ],
696 |       "metadata": {
697 |         "id": "90tlPMFaK-4r"
698 |       },
699 |       "execution_count": null,
700 |       "outputs": []
701 |     },
702 |     {
703 |       "cell_type": "code",
704 |       "source": [
705 |         "model_vit_sup = Sequential([\n",
706 |         "    vit_model,\n",
707 |         "    nn.Dense(2,activation='softmax')\n",
708 |         "])"
709 |       ],
710 |       "metadata": {
711 |         "id": "P7mCIXM-NYVi"
712 |       },
713 |       "execution_count": null,
714 |       "outputs": []
715 |     },
716 |     {
717 |       "cell_type": "code",
718 |       "source": [
719 |         "model_vit_sup.compile(loss=BinaryFocalLoss(gamma=2),optimizer='adam',metrics='accuracy')"
720 |       ],
721 |       "metadata": {
722 |         "id": "po70HSt3NYVj"
723 |       },
724 |       "execution_count": null,
725 |       "outputs": []
726 |     },
727 |     {
728 |       "cell_type": "code",
729 |       "source": [
730 |         "model_vit_sup.fit(Xx,y,epochs=5,verbose=1, batch_size=32)"
731 |       ],
732 |       "metadata": {
733 |         "id": "IFuNQ1-zNYVx"
734 |       },
735 |       "execution_count": null,
736 |       "outputs": []
737 |     },
738 |     {
739 |       "cell_type": "code",
740 |       "source": [
741 |         "model_cnn_sup.save('cnn.keras')\n",
742 |         "model_vit_sup.save('vit.keras')"
743 |       ],
744 |       "metadata": {
745 |         "colab": {
746 |           "base_uri": "https://localhost:8080/"
747 |         },
748 |         "id": "6MSYBg_tlajr",
749 |         "outputId": "23299a17-4f56-4965-c708-2d5476dde9cf"
750 |       },
751 |       "execution_count": null,
752 |       "outputs": [
753 |         {
754 |           "output_type": "stream",
755 |           "name": "stderr",
756 |           "text": [
757 |             "WARNING:absl:Found untraced functions such as _update_step_xla, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op, _jit_compiled_convolution_op while saving (showing 5 of 54). These functions will not be directly callable after loading.\n",
758 |             "WARNING:absl:Function `_wrapped_model` contains input name(s) vit-b32_input with unsupported characters which will be renamed to vit_b32_input in the SavedModel.\n",
759 |             "WARNING:absl:`vit-b32_input` is not a valid tf.function parameter name. Sanitizing to `vit_b32_input`.\n",
760 |             "WARNING:absl:`vit-b32_input` is not a valid tf.function parameter name. Sanitizing to `vit_b32_input`.\n",
761 |             "WARNING:absl:`vit-b32_input` is not a valid tf.function parameter name. Sanitizing to `vit_b32_input`.\n",
762 |             "WARNING:absl:Found untraced functions such as _update_step_xla, _jit_compiled_convolution_op, MultiHeadDotProductAttention_1_layer_call_fn, MultiHeadDotProductAttention_1_layer_call_and_return_conditional_losses, LayerNorm_0_layer_call_fn while saving (showing 5 of 194). These functions will not be directly callable after loading.\n"
763 |           ]
764 |         }
765 |       ]
766 |     },
767 |     {
768 |       "cell_type": "code",
769 |       "source": [
770 |         "import keras_core as keras\n",
771 |         "cnn = keras.saving.load_model('cnn.keras')"
772 |       ],
773 |       "metadata": {
774 |         "id": "dCuOaFFDniIc"
775 |       },
776 |       "execution_count": null,
777 |       "outputs": []
778 |     },
779 |     {
780 |       "cell_type": "code",
781 |       "source": [
782 |         "import keras_core as keras\n",
783 |         "vit = keras.saving.load_model('vit.keras')"
784 |       ],
785 |       "metadata": {
786 |         "id": "A0p-tnxspC23"
787 |       },
788 |       "execution_count": null,
789 |       "outputs": []
790 |     },
791 |     {
792 |       "cell_type": "code",
793 |       "source": [
794 |         "cnn.predict(Xx)"
795 |       ],
796 |       "metadata": {
797 |         "id": "dUXDVmbupHLG"
798 |       },
799 |       "execution_count": null,
800 |       "outputs": []
801 |     },
802 |     {
803 |       "cell_type": "code",
804 |       "source": [
805 |         "!pip install gradio"
806 |       ],
807 |       "metadata": {
808 |         "id": "CoMA4buTpJwO"
809 |       },
810 |       "execution_count": null,
811 |       "outputs": []
812 |     },
813 |     {
814 |       "cell_type": "markdown",
815 |       "source": [
816 |         "## generating the gradio demo for visualisation and hosting"
817 |       ],
818 |       "metadata": {
819 |         "id": "FJ3KwEOGyvb1"
820 |       }
821 |     },
822 |     {
823 |       "cell_type": "code",
824 |       "source": [
825 |         "import gradio as gr\n",
826 |         "import tensorflow as tf\n",
827 |         "from keras_core import Model, Sequential\n",
828 |         "from keras_core.layers import Layer\n",
829 |         "import keras_core.layers as nn\n",
830 |         "import keras_core as keras\n",
831 |         "import numpy as np\n",
832 |         "# vit = keras.models.load_model('vit/')\n",
833 |         "# cnn = keras.models.load_model('cnn/')\n",
834 |         "def fn(image,model_t):\n",
835 |         "  if len(image.shape)==2:\n",
836 |         "    img = np.stack([image,image,image],axis=2)\n",
837 |         "    img = np.resize(img,(224,224,3))\n",
838 |         "  elif len(image.shape)==3 and image.shape[2]==1:\n",
839 |         "    img = np.stack([image[:,:,0],image[:,:,0],image[:,:,0]],axis=2)\n",
840 |         "    img = np.resize(img,(224,224,3))\n",
841 |         "  else:\n",
842 |         "    img = np.resize(image,(224,224,3))\n",
843 |         "  img = np.expand_dims(img,axis=0)\n",
844 |         "  if model_t=='vit':\n",
845 |         "    pred = vit.predict(img)\n",
846 |         "  else:\n",
847 |         "    pred= cnn.predict(img)\n",
848 |         "  if np.argmax(pred)==0:\n",
849 |         "    return 'autism'\n",
850 |         "  else:\n",
851 |         "    return 'control'\n",
852 |         "\n",
853 |         "\n",
854 |         "demo = gr.Interface(\n",
855 |         "    fn,['image',gr.Dropdown(\n",
856 |         "            [\"resnet\", \"vit\"], label=\"model\")],\"text\",\n",
857 |         "\n",
858 |         ")\n",
859 |         "\n",
860 |         "if __name__ == \"__main__\":\n",
861 |         "    demo.launch()"
862 |       ],
863 |       "metadata": {
864 |         "colab": {
865 |           "base_uri": "https://localhost:8080/",
866 |           "height": 706
867 |         },
868 |         "id": "kg2Tlv7irpQA",
869 |         "outputId": "d6c340b0-d108-4579-a0de-bd37f7830389"
870 |       },
871 |       "execution_count": null,
872 |       "outputs": [
873 |         {
874 |           "output_type": "stream",
875 |           "name": "stdout",
876 |           "text": [
877 |             "Colab notebook detected. This cell will run indefinitely so that you can see errors and logs. To turn off, set debug=False in launch().\n",
878 |             "Note: opening Chrome Inspector may crash demo inside Colab notebooks.\n",
879 |             "\n",
880 |             "To create a public link, set `share=True` in `launch()`.\n"
881 |           ]
882 |         },
883 |         {
884 |           "output_type": "display_data",
885 |           "data": {
886 |             "text/plain": [
887 |               "<IPython.core.display.Javascript object>"
888 |             ],
889 |             "application/javascript": [
890 |               "(async (port, path, width, height, cache, element) => {\n",
891 |               "                        if (!google.colab.kernel.accessAllowed && !cache) {\n",
892 |               "                            return;\n",
893 |               "                        }\n",
894 |               "                        element.appendChild(document.createTextNode(''));\n",
895 |               "                        const url = await google.colab.kernel.proxyPort(port, {cache});\n",
896 |               "\n",
897 |               "                        const external_link = document.createElement('div');\n",
898 |               "                        external_link.innerHTML = `\n",
899 |               "                            <div style=\"font-family: monospace; margin-bottom: 0.5rem\">\n",
900 |               "                                Running on <a href=${new URL(path, url).toString()} target=\"_blank\">\n",
901 |               "                                    https://localhost:${port}${path}\n",
902 |               "                                </a>\n",
903 |               "                            </div>\n",
904 |               "                        `;\n",
905 |               "                        element.appendChild(external_link);\n",
906 |               "\n",
907 |               "                        const iframe = document.createElement('iframe');\n",
908 |               "                        iframe.src = new URL(path, url).toString();\n",
909 |               "                        iframe.height = height;\n",
910 |               "                        iframe.allow = \"autoplay; camera; microphone; clipboard-read; clipboard-write;\"\n",
911 |               "                        iframe.width = width;\n",
912 |               "                        iframe.style.border = 0;\n",
913 |               "                        element.appendChild(iframe);\n",
914 |               "                    })(7863, \"/\", \"100%\", 500, false, window.element)"
915 |             ]
916 |           },
917 |           "metadata": {}
918 |         },
919 |         {
920 |           "output_type": "stream",
921 |           "name": "stderr",
922 |           "text": [
923 |             "/usr/local/lib/python3.10/dist-packages/tensorflow/python/data/ops/structured_function.py:254: UserWarning: Even though the `tf.config.experimental_run_functions_eagerly` option is set, this option does not apply to tf.data functions. To force eager execution of tf.data functions, please use `tf.data.experimental.enable_debug_mode()`.\n",
924 |             "  warnings.warn(\n"
925 |           ]
926 |         },
927 |         {
928 |           "output_type": "stream",
929 |           "name": "stdout",
930 |           "text": [
931 |             "1/1 [==============================] - 0s 345ms/step\n",
932 |             "Keyboard interruption in main thread... closing server.\n"
933 |           ]
934 |         }
935 |       ]
936 |     },
937 |     {
938 |       "cell_type": "code",
939 |       "source": [],
940 |       "metadata": {
941 |         "id": "B0FklJxCtbfe"
942 |       },
943 |       "execution_count": null,
944 |       "outputs": []
945 |     }
946 |   ]
947 | }


--------------------------------------------------------------------------------
/README.md:
--------------------------------------------------------------------------------
 1 | ### Problem Statement :
 2 | 
 3 | Early detection of Autism can lead to early intervention which in turn play a huge role in the brain development of children with ASD . This example uses structural and resting state MRI data for Autism Classification .
 4 | 
 5 | ### Model :
 6 | 
 7 | Keras Core and Keras implementation of the paper 
 8 | [CASS: Cross Architectural Self-Supervision for Medical Image Analysis](https://arxiv.org/pdf/2206.04170v6.pdf)
 9 | 
10 | ### Dataset : 
11 | 
12 | ABIDE is a collaboration of 16 international imaging sites that have aggregated and are openly sharing neuroimaging data from 539 individuals suffering from ASD and 573 typical controls. These 1112 datasets are composed of structural and resting state functional MRI data along with an extensive array of phenotypic information.
13 | 
14 | ![ABIDE](https://github.com/ushareng/AI_ForAutism-MRI_ImageClassification/assets/34335028/f9952417-1d18-4c2f-aa9b-bbc757104249)
15 | 
16 | Source : http://preprocessed-connectomes-project.org/abide/
17 | 
18 | Hugging spaces Demo : https://huggingface.co/spaces/tensorgirl/mri_classification
19 | 
20 | ### Intel Extension of TensorFlow
21 | 
22 | Include the line pip install --upgrade intel-extension-for-tensorflow[cpu] in the beginning to use the Intel Extension of Tensorflow
23 | 
24 | ### Reference Papers :
25 | 
26 | https://www.ijrte.org/wp-content/uploads/papers/v7i5s3/E11160275S19.pdf
27 | 
28 | https://scholarworks.uark.edu/cgi/viewcontent.cgi?article=3303&context=jaas
29 | 
30 | ### My Advocacy in Autism
31 | 
32 | #### NeuroAI 
33 | 
34 | https://humansofdata.atlan.com/2019/08/unravel-the-mystery-of-the-human-brain-at-neuroai/
35 | 
36 | #### Neurodiversity India Summit 
37 | 
38 | [2022](https://neuroaiworld.com/neurodiversity-india-summit-2022/)
39 | [2021](https://neuroaiworld.com/neurodiversity-india-summit-2021/)
40 | [2020](https://neuroaiworld.com/neurodiversity-india-summit-2020/)
41 | 
42 | 
43 | 
44 | 
45 | 
46 | 
47 | 
48 | 
49 | 


--------------------------------------------------------------------------------