{
  "id": 18730,
  "title": "Sunnybrook data",
  "url": "/competitions/second-annual-data-science-bowl/discussion/18730",
  "author_name": "",
  "post_date": "2016-02-03T22:51:21.157Z",
  "votes": null,
  "comment_count": 4,
  "views": 1142,
  "content": "<p>Is anyone using the Sunnybrook data? I was thinking about incorporating it but it seems like it doesn't offer very much in terms of data (just 45 images from multiple patients). I was just curious if anyone is using it and if so what their results are. Thanks.</p>",
  "messages": [
    {
      "id": "106791",
      "postDate": "02/03/2016 22:51:21",
      "content": "<p>Is anyone using the Sunnybrook data? I was thinking about incorporating it but it seems like it doesn't offer very much in terms of data (just 45 images from multiple patients). I was just curious if anyone is using it and if so what their results are. Thanks.</p>",
      "rawMarkdown": "Is anyone using the Sunnybrook data? I was thinking about incorporating it but it seems like it doesn't offer very much in terms of data (just 45 images from multiple patients). I was just curious if anyone is using it and if so what their results are. Thanks.",
      "votes": null
    },
    {
      "id": "107555",
      "postDate": "02/11/2016 05:23:31",
      "content": "<p>I took a brief look at it, but it doesn't seem to include the variety of the kaggle set. Lowest age appears to be 23 years old. While it does have 'ground truth', I think a person could ground truth the kaggle training set and get a lot more cases maybe using the Sunnybrook as an example on how to draw the contours. </p>",
      "rawMarkdown": "I took a brief look at it, but it doesn't seem to include the variety of the kaggle set. Lowest age appears to be 23 years old. While it does have 'ground truth', I think a person could ground truth the kaggle training set and get a lot more cases maybe using the Sunnybrook as an example on how to draw the contours.",
      "votes": null
    },
    {
      "id": "107601",
      "postDate": "02/11/2016 16:39:30",
      "content": "<p>That's a good point, I didn't think about using the training set like that. Thanks for the idea.</p>",
      "rawMarkdown": "That's a good point, I didn't think about using the training set like that. Thanks for the idea.",
      "votes": null
    },
    {
      "id": "107808",
      "postDate": "02/12/2016 19:47:37",
      "content": "<p>Are we allowed to do that? If you manually label the data, your script won't run independently anymore ...</p>",
      "rawMarkdown": "Are we allowed to do that? If you manually label the data, your script won't run independently anymore ...",
      "votes": null
    },
    {
      "id": "108528",
      "postDate": "02/18/2016 03:41:29",
      "content": "<p>Did anyone produced labels for training set similar to Sunnybrook?</p>",
      "rawMarkdown": "Did anyone produced labels for training set similar to Sunnybrook?",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 107555,
      "author_name": "darthdavediode",
      "author_url": "",
      "post_date": "02/11/2016 05:23:31",
      "content": "<p>I took a brief look at it, but it doesn't seem to include the variety of the kaggle set. Lowest age appears to be 23 years old. While it does have 'ground truth', I think a person could ground truth the kaggle training set and get a lot more cases maybe using the Sunnybrook as an example on how to draw the contours. </p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 107601,
      "author_name": "patruff",
      "author_url": "",
      "post_date": "02/11/2016 16:39:30",
      "content": "<p>That's a good point, I didn't think about using the training set like that. Thanks for the idea.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 107808,
      "author_name": "wonderingalice",
      "author_url": "",
      "post_date": "02/12/2016 19:47:37",
      "content": "<p>Are we allowed to do that? If you manually label the data, your script won't run independently anymore ...</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 108528,
      "author_name": "serhiy",
      "author_url": "",
      "post_date": "02/18/2016 03:41:29",
      "content": "<p>Did anyone produced labels for training set similar to Sunnybrook?</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "106791": "Is anyone using the Sunnybrook data? I was thinking about incorporating it but it seems like it doesn't offer very much in terms of data (just 45 images from multiple patients). I was just curious if anyone is using it and if so what their results are. Thanks.",
    "107555": "I took a brief look at it, but it doesn't seem to include the variety of the kaggle set. Lowest age appears to be 23 years old. While it does have 'ground truth', I think a person could ground truth the kaggle training set and get a lot more cases maybe using the Sunnybrook as an example on how to draw the contours.",
    "107601": "That's a good point, I didn't think about using the training set like that. Thanks for the idea.",
    "107808": "Are we allowed to do that? If you manually label the data, your script won't run independently anymore ...",
    "108528": "Did anyone produced labels for training set similar to Sunnybrook?"
  },
  "source": "meta"
}