{
  "id": 247868,
  "title": "DICOM full range pixels as CNN input",
  "url": "/competitions/siim-covid19-detection/discussion/247868",
  "author_name": "",
  "post_date": "2021-06-21T13:38:23.608241300Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>I made this notebook a few days ago and just realized I didn't make it public.</p>\n<p>This is a simple example to show how to use the full width of DICOM pixels as input for CNN .. rather than exporting 8 bit JPG. In this case, I feed the images to a pretrained InceptionV3 NN for transfer learning.</p>\n<p><a href=\"https://www.kaggle.com/davidbroberts/dicom-full-range-pixels-as-cnn-input\" target=\"_blank\">https://www.kaggle.com/davidbroberts/dicom-full-range-pixels-as-cnn-input</a></p>",
  "messages": [
    {
      "id": "1359737",
      "postDate": "06/21/2021 13:38:23",
      "content": "<p>I made this notebook a few days ago and just realized I didn't make it public.</p>\n<p>This is a simple example to show how to use the full width of DICOM pixels as input for CNN .. rather than exporting 8 bit JPG. In this case, I feed the images to a pretrained InceptionV3 NN for transfer learning.</p>\n<p><a href=\"https://www.kaggle.com/davidbroberts/dicom-full-range-pixels-as-cnn-input\" target=\"_blank\">https://www.kaggle.com/davidbroberts/dicom-full-range-pixels-as-cnn-input</a></p>",
      "rawMarkdown": "I made this notebook a few days ago and just realized I didn't make it public.\n\nThis is a simple example to show how to use the full width of DICOM pixels as input for CNN .. rather than exporting 8 bit JPG. In this case, I feed the images to a pretrained InceptionV3 NN for transfer learning.\n\nhttps://www.kaggle.com/davidbroberts/dicom-full-range-pixels-as-cnn-input",
      "votes": null
    },
    {
      "id": "1361401",
      "postDate": "06/22/2021 19:25:39",
      "content": "<p>Now I know the importance of discussion. <br>\nNice work sir.</p>",
      "rawMarkdown": "Now I know the importance of discussion. \nNice work sir.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1361401,
      "author_name": "prateek147",
      "author_url": "",
      "post_date": "06/22/2021 19:25:39",
      "content": "<p>Now I know the importance of discussion. <br>\nNice work sir.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1359737": "I made this notebook a few days ago and just realized I didn't make it public.\n\nThis is a simple example to show how to use the full width of DICOM pixels as input for CNN .. rather than exporting 8 bit JPG. In this case, I feed the images to a pretrained InceptionV3 NN for transfer learning.\n\nhttps://www.kaggle.com/davidbroberts/dicom-full-range-pixels-as-cnn-input",
    "1361401": "Now I know the importance of discussion. \nNice work sir."
  },
  "source": "meta"
}