{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\n\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        os.path.join(dirname, filename)\n\n# You can write up to 5GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","execution_count":null,"outputs":[]},{"metadata":{"_uuid":"d629ff2d2480ee46fbb7e2d37f6b5fab8052498a","_cell_guid":"79c7e3d0-c299-4dcb-8224-4455121ee9b0","trusted":true},"cell_type":"code","source":"import matplotlib.pyplot as plt\nimport pydicom\nfrom pydicom.data import get_testdata_files\nimport glob as glob\n\npath = \"../input/osic-pulmonary-fibrosis-progression/train/\"\npath1 = \"../input/osic-pulmonary-fibrosis-progression/train/ID00007637202177411956430/19.dcm\"\npath_patient1 = \"../input/osic-pulmonary-fibrosis-progression/train/ID00007637202177411956430/\"\nimg1 = pydicom.dcmread(path1)\n\nprint(img1.pixel_array.shape)\nplt.figure(figsize = (7, 7))\nplt.imshow(img1.pixel_array, cmap=\"plasma\")\nplt.axis('off');","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"print(\"Patient id.......:\", img1.PatientID, \"\\n\" +\n      \"Modality.........:\", img1.Modality, \"\\n\" +\n      \"Rows.............:\", img1.Rows, \"\\n\" +\n      \"Columns..........:\", img1.Columns)\n\nprint(\"img1:\", img1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"data_path = '../input/osic-pulmonary-fibrosis-progression/train/'\n\noutput_path = '../input/output/'\ntrain_image_files = sorted(glob.glob(os.path.join(data_path, '*','*.dcm')))\npatients = os.listdir(data_path)\npatients.sort()\n\nprint('Some sample Patient ID''s :', len(train_image_files))\nprint(\"\\n\".join(train_image_files[:5]))\n","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"In the following lines of code I tried to create baseline CNN which takes 10 photos of 1 patient, and give his mean FVC (taken from the tabular data) as output."},{"metadata":{"trusted":true},"cell_type":"code","source":"def load_scan(path):\n    slices = [pydicom.read_file(s) for s in path[0:10]]\n    return slices","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"data_10_photos = load_scan(train_image_files)\nimg_data = [img.pixel_array for img in data_10_photos]\nprint(img_data)\nprint(img_data[0][100]) # the 100th raw in the first img","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"ids = [img.PatientID for img in data_10_photos]\nids","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# import tabular data\n\ntabular_dataset_train = pd.read_csv(\"../input/osic-pulmonary-fibrosis-progression/train.csv\")\ntabular_dataset_train.head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# create new df with patient_id and mean_fvc \n\nnew_df = tabular_dataset_train.groupby('Patient').mean('FVC').drop(columns = ['Age', 'Percent', 'Weeks'])\nprint(new_df)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import mean_absolute_error\nfrom tensorflow.python import keras\nfrom tensorflow.python.keras.models import Sequential\nfrom tensorflow.python.keras.layers import Dense, Flatten, Conv2D, Dropout, MaxPooling2D\n\n\nimg_rows, img_cols = 512, 512\nnum_images = 10\n\nx = np.array([img_data]).reshape(num_images, img_rows, img_cols, 1)\ny = np.array([2113 for i in range(10)]) # mean fvc of that patient\n    \nmodel = Sequential()\nmodel.add(Conv2D(20, kernel_size=(3, 3),\n                 activation='relu',\n                 input_shape=(img_rows, img_cols, 1)))\nmodel.add(MaxPooling2D(pool_size=(3, 3)))\nmodel.add(Conv2D(20, kernel_size=(3, 3), activation='relu'))\nmodel.add(Flatten())\nmodel.add(Dense(128, activation='relu'))\nmodel.add(Dense(1, activation='linear'))\n\nmodel.compile(loss='mean_absolute_error', optimizer='adam', metrics=['mean_absolute_error'])\n\nmodel.summary()\n\nhistory = model.fit(x, y,\n          batch_size=1,\n          epochs=30,\n          validation_split = 0.2)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"mae_train = history.history['mean_absolute_error']\nmae_val = history.history['val_mean_absolute_error']\nepochs = range(1,31)\nplt.plot(epochs, mae_train, 'g', label='Training mae')\nplt.plot(epochs, mae_val, 'b', label='validation mae')\nplt.title('Training and Validation Mean Absolure Error')\nplt.xlabel('Epochs')\nplt.ylabel('MAE')\nplt.legend()\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"mae_train = history.history['mean_absolute_error']\nmae_val = history.history['val_mean_absolute_error']\nepochs = range(1,31)\nplt.plot(epochs, mae_train, 'g', label='Training mae')\nplt.plot(epochs, mae_val, 'b', label='validation mae')\nplt.title('Training and Validation Mean Absolure Error (log scale)')\nplt.xlabel('Epochs')\nplt.ylabel('MAE')\nplt.yscale(\"log\")\nplt.legend()\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def load_scan(path):\n    slices = [pydicom.read_file(path + '/' + s) for s in os.listdir(path)]\n    return slices","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# import all CT scans\n\nfrom pathlib import Path\nroot_dir = Path('/kaggle/input/osic-pulmonary-fibrosis-progression/train')\ndef load_scan(path):\n    slices = [pydicom.read_file(p) for p in path.glob('*.dcm')]\n    image = np.stack([s.pixel_array.astype(float) for s in slices])\n    return image, slices[0]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# type(slices[0])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# add FVC values to scans data","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# create CNN model with: X = scans, y = FVC","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat":4,"nbformat_minor":4}