{
  "id": 109531,
  "title": "DCM metadata usage question.",
  "url": "/competitions/rsna-intracranial-hemorrhage-detection/discussion/109531",
  "author_name": "Daryl Brach",
  "post_date": "2019-09-20T00:52:26.560000",
  "votes": 0,
  "comment_count": 4,
  "views": 0,
  "content": "<p>I understand that \"Submission predictions must be based entirely on the pixel data in the provided datasets.\" However, does that include pixel related data in the metadata or not. Specifically, I would like to know if we can use the WindowCenter, WindowWidth, RescaleIntercep, and RescaleSlope data from the DCM file to accurately convert the pixel value to Hundsfeld units.</p>",
  "messages": [
    {
      "id": 643385,
      "postDate": "2019-10-07T13:35:21.707Z",
      "content": "<p><strong>what's the meaning of Hundsfeld units?</strong></p>",
      "rawMarkdown": "**what's the meaning of Hundsfeld units?**"
    },
    {
      "id": 630846,
      "postDate": "2019-09-20T20:51:14.600Z",
      "content": "<p>Hi!  Thanks for your questions.  A similar question was asked <a href=\"https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/109479#latest-630263\">here</a> and the host's answer was that that sort of use of the metadata is indeed acceptable.</p>",
      "rawMarkdown": "Hi!  Thanks for your questions.  A similar question was asked [here](https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/109479#latest-630263) and the host's answer was that that sort of use of the metadata is indeed acceptable.",
      "replies": [
        {
          "id": 630860,
          "postDate": "2019-09-20T21:50:43.870Z",
          "content": "<p>Thanks Phil!</p>",
          "rawMarkdown": "Thanks Phil!"
        }
      ]
    },
    {
      "id": 630631,
      "postDate": "2019-09-20T14:23:30.473Z",
      "content": "<p>I have the same question. Actually, in this dataset, the dcm files have different PixelSpacing. Can we use this metadata to normalize the images?👀 </p>",
      "rawMarkdown": "I have the same question. Actually, in this dataset, the dcm files have different PixelSpacing. Can we use this metadata to normalize the images?👀 "
    },
    {
      "id": 630233,
      "postDate": "2019-09-20T00:52:26.560Z",
      "content": "<p>I understand that \"Submission predictions must be based entirely on the pixel data in the provided datasets.\" However, does that include pixel related data in the metadata or not. Specifically, I would like to know if we can use the WindowCenter, WindowWidth, RescaleIntercep, and RescaleSlope data from the DCM file to accurately convert the pixel value to Hundsfeld units.</p>",
      "rawMarkdown": "I understand that \"Submission predictions must be based entirely on the pixel data in the provided datasets.\" However, does that include pixel related data in the metadata or not. Specifically, I would like to know if we can use the WindowCenter, WindowWidth, RescaleIntercep, and RescaleSlope data from the DCM file to accurately convert the pixel value to Hundsfeld units."
    }
  ],
  "comments": [
    {
      "id": 643385,
      "author_name": "pupil3",
      "author_url": "",
      "post_date": "2019-10-07T13:35:21.707000",
      "content": "<p><strong>what's the meaning of Hundsfeld units?</strong></p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 630846,
      "author_name": "Phil Culliton",
      "author_url": "",
      "post_date": "2019-09-20T20:51:14.600000",
      "content": "<p>Hi!  Thanks for your questions.  A similar question was asked <a href=\"https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/109479#latest-630263\">here</a> and the host's answer was that that sort of use of the metadata is indeed acceptable.</p>",
      "votes": 0,
      "replies": [
        {
          "id": 630860,
          "author_name": "Daryl Brach",
          "author_url": "",
          "post_date": "2019-09-20T21:50:43.870000",
          "content": "<p>Thanks Phil!</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 630631,
      "author_name": "wangyunpeng_bio",
      "author_url": "",
      "post_date": "2019-09-20T14:23:30.473000",
      "content": "<p>I have the same question. Actually, in this dataset, the dcm files have different PixelSpacing. Can we use this metadata to normalize the images?👀 </p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "643385": "**what's the meaning of Hundsfeld units?**",
    "630846": "Hi!  Thanks for your questions.  A similar question was asked [here](https://www.kaggle.com/c/rsna-intracranial-hemorrhage-detection/discussion/109479#latest-630263) and the host's answer was that that sort of use of the metadata is indeed acceptable.",
    "630631": "I have the same question. Actually, in this dataset, the dcm files have different PixelSpacing. Can we use this metadata to normalize the images?👀 ",
    "630233": "I understand that \"Submission predictions must be based entirely on the pixel data in the provided datasets.\" However, does that include pixel related data in the metadata or not. Specifically, I would like to know if we can use the WindowCenter, WindowWidth, RescaleIntercep, and RescaleSlope data from the DCM file to accurately convert the pixel value to Hundsfeld units."
  }
}