Age and Contrast prediction

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":2343,"sourceType":"datasetVersion","datasetId":1012}],"dockerImageVersionId":30558,"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"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        print(os.path.join(dirname, filename))\n\n# You can write up to 20GB 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","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2024-06-17T06:54:11.597409Z","iopub.execute_input":"2024-06-17T06:54:11.597857Z","iopub.status.idle":"2024-06-17T06:54:11.620457Z","shell.execute_reply.started":"2024-06-17T06:54:11.597822Z","shell.execute_reply":"2024-06-17T06:54:11.618871Z"},"trusted":true},"execution_count":13,"outputs":[{"name":"stdout","text":"/kaggle/input/siim-medical-images/full_archive.npz\n/kaggle/input/siim-medical-images/overview.csv\n/kaggle/input/siim-medical-images/dicom_dir/ID_0015_AGE_0061_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0019_AGE_0070_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0063_AGE_0073_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0077_AGE_0074_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0073_AGE_0074_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0060_AGE_0080_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0051_AGE_0063_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0091_AGE_0072_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0004_AGE_0056_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0076_AGE_0068_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0055_AGE_0071_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0000_AGE_0060_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0089_AGE_0071_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0036_AGE_0074_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0066_AGE_0082_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0078_AGE_0066_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0064_AGE_0058_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0030_AGE_0076_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0016_AGE_0063_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0052_AGE_0072_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0081_AGE_0058_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0041_AGE_0045_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0014_AGE_0071_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0001_AGE_0069_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0093_AGE_0067_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0012_AGE_0061_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0097_AGE_0060_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0021_AGE_0067_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0023_AGE_0061_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0031_AGE_0039_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0049_AGE_0061_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0046_AGE_0072_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0065_AGE_0082_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0040_AGE_0069_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0088_AGE_0067_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0075_AGE_0080_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0053_AGE_0073_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0054_AGE_0082_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0032_AGE_0061_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0026_AGE_0070_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0083_AGE_0082_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0085_AGE_0067_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0074_AGE_0074_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0034_AGE_0061_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0013_AGE_0060_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0057_AGE_0049_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0011_AGE_0061_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0038_AGE_0071_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0010_AGE_0060_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0098_AGE_0061_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0094_AGE_0052_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0039_AGE_0074_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0047_AGE_0069_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0082_AGE_0047_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0037_AGE_0074_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0050_AGE_0074_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0006_AGE_0075_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0096_AGE_0083_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0048_AGE_0077_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0099_AGE_0061_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0059_AGE_0074_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0092_AGE_0072_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0080_AGE_0070_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0008_AGE_0051_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0095_AGE_0071_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0070_AGE_0074_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0045_AGE_0074_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0003_AGE_0075_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0002_AGE_0074_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0025_AGE_0074_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0007_AGE_0061_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0071_AGE_0065_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0084_AGE_0067_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0029_AGE_0078_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0061_AGE_0074_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0018_AGE_0074_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0017_AGE_0060_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0043_AGE_0069_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0056_AGE_0083_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0005_AGE_0048_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0020_AGE_0066_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0079_AGE_0071_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0068_AGE_0072_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0069_AGE_0074_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0062_AGE_0067_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0086_AGE_0073_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0087_AGE_0044_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0044_AGE_0072_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0022_AGE_0074_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0027_AGE_0064_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0009_AGE_0048_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0090_AGE_0067_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0028_AGE_0074_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0033_AGE_0071_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0035_AGE_0059_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0058_AGE_0082_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0072_AGE_0060_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0067_AGE_0060_CONTRAST_0_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0042_AGE_0071_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/dicom_dir/ID_0024_AGE_0060_CONTRAST_1_CT.dcm\n/kaggle/input/siim-medical-images/tiff_images/ID_0027_AGE_0064_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0096_AGE_0083_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0073_AGE_0074_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0046_AGE_0072_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0052_AGE_0072_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0015_AGE_0061_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0009_AGE_0048_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0048_AGE_0077_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0020_AGE_0066_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0055_AGE_0071_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0077_AGE_0074_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0086_AGE_0073_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0005_AGE_0048_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0050_AGE_0074_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0002_AGE_0074_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0016_AGE_0063_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0083_AGE_0082_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0063_AGE_0073_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0038_AGE_0071_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0024_AGE_0060_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0023_AGE_0061_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0019_AGE_0070_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0072_AGE_0060_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0074_AGE_0074_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0076_AGE_0068_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0001_AGE_0069_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0061_AGE_0074_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0035_AGE_0059_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0070_AGE_0074_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0095_AGE_0071_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0047_AGE_0069_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0022_AGE_0074_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0094_AGE_0052_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0029_AGE_0078_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0057_AGE_0049_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0056_AGE_0083_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0064_AGE_0058_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0079_AGE_0071_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0011_AGE_0061_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0033_AGE_0071_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0039_AGE_0074_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0067_AGE_0060_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0065_AGE_0082_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0087_AGE_0044_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0066_AGE_0082_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0017_AGE_0060_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0053_AGE_0073_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0040_AGE_0069_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0021_AGE_0067_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0010_AGE_0060_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0062_AGE_0067_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0025_AGE_0074_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0004_AGE_0056_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0090_AGE_0067_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0080_AGE_0070_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0060_AGE_0080_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0089_AGE_0071_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0098_AGE_0061_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0014_AGE_0071_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0082_AGE_0047_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0007_AGE_0061_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0059_AGE_0074_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0012_AGE_0061_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0041_AGE_0045_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0068_AGE_0072_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0097_AGE_0060_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0054_AGE_0082_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0071_AGE_0065_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0036_AGE_0074_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0030_AGE_0076_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0091_AGE_0072_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0078_AGE_0066_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0069_AGE_0074_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0037_AGE_0074_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0099_AGE_0061_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0045_AGE_0074_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0034_AGE_0061_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0043_AGE_0069_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0085_AGE_0067_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0049_AGE_0061_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0093_AGE_0067_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0000_AGE_0060_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0084_AGE_0067_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0058_AGE_0082_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0013_AGE_0060_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0088_AGE_0067_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0042_AGE_0071_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0092_AGE_0072_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0018_AGE_0074_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0051_AGE_0063_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0031_AGE_0039_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0006_AGE_0075_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0032_AGE_0061_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0008_AGE_0051_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0044_AGE_0072_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0026_AGE_0070_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0003_AGE_0075_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0028_AGE_0074_CONTRAST_1_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0075_AGE_0080_CONTRAST_0_CT.tif\n/kaggle/input/siim-medical-images/tiff_images/ID_0081_AGE_0058_CONTRAST_0_CT.tif\n","output_type":"stream"}]},{"cell_type":"code","source":"pip install pydicom","metadata":{"execution":{"iopub.status.busy":"2024-06-17T06:54:11.623036Z","iopub.execute_input":"2024-06-17T06:54:11.623463Z","iopub.status.idle":"2024-06-17T06:54:46.207712Z","shell.execute_reply.started":"2024-06-17T06:54:11.623426Z","shell.execute_reply":"2024-06-17T06:54:46.206223Z"},"trusted":true},"execution_count":14,"outputs":[{"name":"stdout","text":"Requirement already satisfied: pydicom in /opt/conda/lib/python3.10/site-packages (2.4.3)\nNote: you may need to restart the kernel to use updated packages.\n","output_type":"stream"}]},{"cell_type":"code","source":"import pandas as pd\n\n# Load the CSV file\ndf = pd.read_csv('/kaggle/input/siim-medical-images/overview.csv')\n\n# Print the column names to check for 'image_filename'\nprint(df.columns)","metadata":{"execution":{"iopub.status.busy":"2024-06-17T06:54:46.209553Z","iopub.execute_input":"2024-06-17T06:54:46.209953Z","iopub.status.idle":"2024-06-17T06:54:46.230631Z","shell.execute_reply.started":"2024-06-17T06:54:46.209921Z","shell.execute_reply":"2024-06-17T06:54:46.229078Z"},"trusted":true},"execution_count":15,"outputs":[{"name":"stdout","text":"Index(['Unnamed: 0', 'Age', 'Contrast', 'ContrastTag', 'raw_input_path', 'id',\n       'tiff_name', 'dicom_name'],\n      dtype='object')\n","output_type":"stream"}]},{"cell_type":"code","source":"import os\nimport pandas as pd\nimport numpy as np\nimport tensorflow as tf\nfrom sklearn.preprocessing import LabelEncoder\nfrom sklearn.model_selection import train_test_split\nfrom tensorflow import keras\nfrom tensorflow.keras import layers\nimport pydicom \n\n# Load the dataset\ndata_dir = '/kaggle/input/siim-medical-images'\ndf = pd.read_csv(os.path.join(data_dir, 'overview.csv'))\n\n# Preprocess the labels\ny_contrast = LabelEncoder().fit_transform(df['Contrast'])\ny_age = df['Age']\n\n# Load and preprocess images\nimage_data = []\ndicom_image_folder = os.path.join(data_dir, 'dicom_dir')\n\nfor filename in df['dicom_name']:\n    image_path = os.path.join(dicom_image_folder, filename)\n\n    if os.path.exists(image_path):\n        dicom_data = pydicom.dcmread(image_path)\n        image = dicom_data.pixel_array\n        image = tf.image.convert_image_dtype(image, tf.float32)\n\n        # Add channel dimension for single-channel images\n        image = tf.expand_dims(image, axis=-1)\n\n        # Resize the image to (224, 224)\n        image = tf.image.resize(image, (224, 224))\n\n        image_data.append(image)\n    else:\n        print(f\"File not found: {image_path}\")\n\n\nX = np.array(image_data)\n\n# Split the data into training and testing sets\nX_train, X_test, y_contrast_train, y_contrast_test, y_age_train, y_age_test = train_test_split(\n    X, y_contrast, y_age, test_size=0.2, random_state=42)\n\n# Build a CNN model for contrast prediction\nmodel_contrast = keras.Sequential([\n    layers.Input(shape=(224, 224, 1)),\n    layers.Conv2D(32, (3, 3), activation='relu'),\n    layers.MaxPooling2D((2, 2)),\n    layers.Conv2D(64, (3, 3), activation='relu'),\n    layers.MaxPooling2D((2, 2)),\n    layers.Conv2D(64, (3, 3), activation='relu'),\n    layers.Flatten(),\n    layers.Dense(64, activation='relu'),\n    layers.Dense(2, activation='softmax')  # 2 classes: with and without contrast\n])\n\n# Compile the model for contrast prediction\nmodel_contrast.compile(optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['accuracy'])\n\n# Train the model for contrast prediction\nhistory_contrast = model_contrast.fit(X_train, y_contrast_train, epochs=10, validation_split=0.2)\nmodel_contrast.fit(X_train, y_contrast_train, epochs=10, validation_split=0.2)\n# Build a CNN model for age prediction\nmodel_age = keras.Sequential([\n    layers.Input(shape=(224, 224, 1)),\n    layers.Conv2D(32, (3, 3), activation='relu'),\n    layers.MaxPooling2D((2, 2)),\n    layers.Conv2D(64, (3, 3), activation='relu'),\n    layers.MaxPooling2D((2, 2)),\n    layers.Conv2D(64, (3, 3), activation='relu'),\n    layers.Flatten(),\n    layers.Dense(64, activation='relu'),\n    layers.Dense(1)  # Regression for age prediction\n])\n\n# Compile the model for age prediction\nmodel_age.compile(optimizer='adam', loss='mean_squared_error', metrics=['mae'])\n\n# Train the model for age prediction\nhistory_age = model_age.fit(X_train, y_age_train, epochs=10, validation_split=0.2)\nmodel_age.fit(X_train, y_age_train, epochs=10, validation_split=0.2)\n\n# Predict contrast usage on the test set\ny_contrast_pred = model_contrast.predict(X_test)\n\n# Predict age on the test set\ny_age_pred = model_age.predict(X_test)\n","metadata":{"execution":{"iopub.status.busy":"2024-06-17T06:54:46.234866Z","iopub.execute_input":"2024-06-17T06:54:46.235450Z","iopub.status.idle":"2024-06-17T06:57:26.764892Z","shell.execute_reply.started":"2024-06-17T06:54:46.235406Z","shell.execute_reply":"2024-06-17T06:57:26.763409Z"},"trusted":true},"execution_count":16,"outputs":[{"name":"stdout","text":"Epoch 1/10\n2/2 [==============================] - 5s 2s/step - loss: 0.7604 - accuracy: 0.5000 - val_loss: 0.6904 - val_accuracy: 0.6250\nEpoch 2/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6943 - accuracy: 0.5469 - val_loss: 0.6971 - val_accuracy: 0.3750\nEpoch 3/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6931 - accuracy: 0.5000 - val_loss: 0.6959 - val_accuracy: 0.3750\nEpoch 4/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6934 - accuracy: 0.5000 - val_loss: 0.6941 - val_accuracy: 0.3750\nEpoch 5/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6929 - accuracy: 0.5000 - val_loss: 0.6951 - val_accuracy: 0.3750\nEpoch 6/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6926 - accuracy: 0.5000 - val_loss: 0.6969 - val_accuracy: 0.3750\nEpoch 7/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6932 - accuracy: 0.5000 - val_loss: 0.6980 - val_accuracy: 0.3750\nEpoch 8/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6941 - accuracy: 0.5000 - val_loss: 0.6943 - val_accuracy: 0.3750\nEpoch 9/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6918 - accuracy: 0.5000 - val_loss: 0.6938 - val_accuracy: 0.3750\nEpoch 10/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6910 - accuracy: 0.5000 - val_loss: 0.6947 - val_accuracy: 0.3750\nEpoch 1/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6900 - accuracy: 0.5000 - val_loss: 0.6957 - val_accuracy: 0.3750\nEpoch 2/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6876 - accuracy: 0.5000 - val_loss: 0.6939 - val_accuracy: 0.3750\nEpoch 3/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6864 - accuracy: 0.5000 - val_loss: 0.6927 - val_accuracy: 0.3750\nEpoch 4/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6827 - accuracy: 0.5000 - val_loss: 0.6924 - val_accuracy: 0.3750\nEpoch 5/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6789 - accuracy: 0.5000 - val_loss: 0.6946 - val_accuracy: 0.3750\nEpoch 6/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6772 - accuracy: 0.5000 - val_loss: 0.6896 - val_accuracy: 0.3750\nEpoch 7/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6770 - accuracy: 0.5156 - val_loss: 0.6867 - val_accuracy: 0.6250\nEpoch 8/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6736 - accuracy: 0.6250 - val_loss: 0.6850 - val_accuracy: 0.6250\nEpoch 9/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6699 - accuracy: 0.5781 - val_loss: 0.6906 - val_accuracy: 0.3750\nEpoch 10/10\n2/2 [==============================] - 4s 2s/step - loss: 0.6681 - accuracy: 0.5000 - val_loss: 0.6811 - val_accuracy: 0.4375\nEpoch 1/10\n2/2 [==============================] - 5s 2s/step - loss: 4570.5894 - mae: 66.9730 - val_loss: 4162.1963 - val_mae: 63.8796\nEpoch 2/10\n2/2 [==============================] - 4s 2s/step - loss: 3879.8252 - mae: 61.5115 - val_loss: 2256.0615 - val_mae: 46.7446\nEpoch 3/10\n2/2 [==============================] - 4s 2s/step - loss: 1567.2134 - mae: 37.0099 - val_loss: 208.0033 - val_mae: 11.7702\nEpoch 4/10\n2/2 [==============================] - 4s 2s/step - loss: 946.5182 - mae: 25.4094 - val_loss: 873.7595 - val_mae: 28.2817\nEpoch 5/10\n2/2 [==============================] - 4s 2s/step - loss: 465.6400 - mae: 18.2363 - val_loss: 138.6269 - val_mae: 10.6754\nEpoch 6/10\n2/2 [==============================] - 4s 2s/step - loss: 297.0953 - mae: 14.8917 - val_loss: 592.9769 - val_mae: 22.9119\nEpoch 7/10\n2/2 [==============================] - 4s 2s/step - loss: 657.8127 - mae: 24.0204 - val_loss: 522.9601 - val_mae: 21.3105\nEpoch 8/10\n2/2 [==============================] - 4s 2s/step - loss: 466.4525 - mae: 19.6546 - val_loss: 153.7738 - val_mae: 11.3720\nEpoch 9/10\n2/2 [==============================] - 4s 2s/step - loss: 130.9128 - mae: 9.6912 - val_loss: 158.6938 - val_mae: 9.7513\nEpoch 10/10\n2/2 [==============================] - 4s 2s/step - loss: 213.9691 - mae: 12.2300 - val_loss: 369.7918 - val_mae: 17.1720\nEpoch 1/10\n2/2 [==============================] - 4s 2s/step - loss: 313.5738 - mae: 15.2860 - val_loss: 143.2644 - val_mae: 9.1246\nEpoch 2/10\n2/2 [==============================] - 4s 2s/step - loss: 107.0546 - mae: 8.5240 - val_loss: 89.6543 - val_mae: 8.0465\nEpoch 3/10\n2/2 [==============================] - 4s 2s/step - loss: 127.7654 - mae: 9.0129 - val_loss: 180.7979 - val_mae: 12.4872\nEpoch 4/10\n2/2 [==============================] - 4s 2s/step - loss: 203.7301 - mae: 12.3240 - val_loss: 154.7383 - val_mae: 11.4803\nEpoch 5/10\n2/2 [==============================] - 4s 2s/step - loss: 150.0586 - mae: 10.0739 - val_loss: 77.6943 - val_mae: 7.0838\nEpoch 6/10\n2/2 [==============================] - 4s 2s/step - loss: 92.3316 - mae: 7.5295 - val_loss: 108.8896 - val_mae: 7.8654\nEpoch 7/10\n2/2 [==============================] - 4s 2s/step - loss: 125.9598 - mae: 8.9476 - val_loss: 143.1233 - val_mae: 9.0938\nEpoch 8/10\n2/2 [==============================] - 4s 2s/step - loss: 129.0930 - mae: 9.2103 - val_loss: 86.7342 - val_mae: 6.7978\nEpoch 9/10\n2/2 [==============================] - 4s 2s/step - loss: 82.8021 - mae: 7.3671 - val_loss: 79.2526 - val_mae: 7.2213\nEpoch 10/10\n2/2 [==============================] - 4s 2s/step - loss: 92.9709 - mae: 7.4852 - val_loss: 98.8538 - val_mae: 8.6280\n1/1 [==============================] - 0s 357ms/step\n1/1 [==============================] - 0s 361ms/step\n","output_type":"stream"}]},{"cell_type":"code","source":"from keras.utils import plot_model\n\n# Generate and save the model architecture diagram\nplot_model(model_contrast, to_file='model_contrast.png', show_shapes=True)\n","metadata":{"execution":{"iopub.status.busy":"2024-06-17T06:57:26.766751Z","iopub.execute_input":"2024-06-17T06:57:26.767212Z","iopub.status.idle":"2024-06-17T06:57:26.888178Z","shell.execute_reply.started":"2024-06-17T06:57:26.767175Z","shell.execute_reply":"2024-06-17T06:57:26.886645Z"},"trusted":true},"execution_count":17,"outputs":[{"execution_count":17,"output_type":"execute_result","data":{"image/png":"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","text/plain":"<IPython.core.display.Image object>"},"metadata":{}}]},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\n# Collect training history\n#history = model_contrast.fit(X_train, y_contrast_train, epochs=10, validation_split=0.2)\n\n# Plot training and validation loss\nplt.plot(history_contrast.history['loss'], label='Training Loss')\nplt.plot(history_contrast.history['val_loss'], label='Validation Loss')\nplt.title('Model Training Curves')\nplt.xlabel('Epochs')\nplt.ylabel('Loss')\nplt.legend()\nplt.show()\n","metadata":{"execution":{"iopub.status.busy":"2024-06-17T06:57:26.889774Z","iopub.execute_input":"2024-06-17T06:57:26.890140Z","iopub.status.idle":"2024-06-17T06:57:27.257259Z","shell.execute_reply.started":"2024-06-17T06:57:26.890109Z","shell.execute_reply":"2024-06-17T06:57:27.256006Z"},"trusted":true},"execution_count":18,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"import random\nimport matplotlib.pyplot as plt\n\n# Select a random sample from the test data\nsample_index = random.randint(0, len(X_test))\nsample_image = X_test[sample_index]\nsample_ground_truth = y_contrast_test[sample_index]\n\n# Get the model's prediction for this sample\nsample_prediction = model_contrast.predict(sample_image.reshape(1, 224, 224, 1))\n\n# Define class labels\nclass_labels = ['Without Contrast', 'With Contrast']\n\n# Create a bar chart to visualize the model's predictions\nplt.figure(figsize=(8, 6))\nplt.bar(class_labels, sample_prediction[0])\nplt.title('Model Predictions for Contrast')\nplt.xlabel('Contrast')\nplt.ylabel('Prediction Probability')\nplt.show()\n","metadata":{"execution":{"iopub.status.busy":"2024-06-17T06:57:27.259099Z","iopub.execute_input":"2024-06-17T06:57:27.259580Z","iopub.status.idle":"2024-06-17T06:57:27.671738Z","shell.execute_reply.started":"2024-06-17T06:57:27.259535Z","shell.execute_reply":"2024-06-17T06:57:27.670527Z"},"trusted":true},"execution_count":19,"outputs":[{"name":"stdout","text":"1/1 [==============================] - 0s 53ms/step\n","output_type":"stream"},{"output_type":"display_data","data":{"text/plain":"<Figure size 800x600 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"from sklearn.metrics import confusion_matrix, classification_report\nimport matplotlib.pyplot as plt\nimport seaborn as sns\n\n# Convert the predicted probabilities to class labels\ny_pred_labels = np.argmax(y_contrast_pred, axis=1)\n\n# Convert the numeric labels in y_contrast_test to strings\ny_contrast_test = [str(label) for label in y_contrast_test]\n\n# Convert the predicted labels to strings\ny_pred_labels = [str(label) for label in y_pred_labels]\n\n# Create a confusion matrix\nconf_matrix = confusion_matrix(y_contrast_test, y_pred_labels)\n\n# Plot the confusion matrix\nplt.figure(figsize=(8, 6))\nsns.heatmap(conf_matrix, annot=True, fmt='d', cmap='Blues')\nplt.title('Confusion Matrix for Contrast Prediction')\nplt.xlabel('Predicted')\nplt.ylabel('Actual')\nplt.show()\n\n# Create a classification report\nprint(classification_report(y_contrast_test, y_pred_labels))\n","metadata":{"execution":{"iopub.status.busy":"2024-06-17T06:57:27.673052Z","iopub.execute_input":"2024-06-17T06:57:27.673425Z","iopub.status.idle":"2024-06-17T06:57:28.049831Z","shell.execute_reply.started":"2024-06-17T06:57:27.673395Z","shell.execute_reply":"2024-06-17T06:57:28.048148Z"},"trusted":true},"execution_count":20,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 800x600 with 2 Axes>","image/png":"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"},"metadata":{}},{"name":"stdout","text":"              precision    recall  f1-score   support\n\n           0       1.00      0.25      0.40         8\n           1       0.67      1.00      0.80        12\n\n    accuracy                           0.70        20\n   macro avg       0.83      0.62      0.60        20\nweighted avg       0.80      0.70      0.64        20\n\n","output_type":"stream"}]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}