{
  "id": 240221,
  "title": "Dataset Containing Additional Information About the Train Dataset (image shape, etc.)",
  "url": "/competitions/siim-covid19-detection/discussion/240221",
  "author_name": "",
  "post_date": "2021-05-18T23:21:23.533129400Z",
  "votes": 8,
  "comment_count": 5,
  "views": 0,
  "content": "<p>Hi there.</p>\n<p>I am working on an <a href=\"https://www.kaggle.com/dschettler8845/visual-in-depth-eda-siim-covid19-detection/output\" target=\"_blank\">EDA</a>, see here, and one of the outcomes is that I ended up with a data-frame/csv containing lots of useful information. I created a <a href=\"https://www.kaggle.com/dschettler8845/siim-covid19-updated-train-labels\" target=\"_blank\">dataset</a> for it to allow for easy access.</p>\n<p>I'm hoping this will save some people some time.</p>\n<p>Some examples:</p>\n<ul>\n<li><strong>Merged the two provided csvs into one on StudyInstanceUID</strong></li>\n<li><strong>Image Shape</strong> (takes a while to compute but now you don't have to)</li>\n<li><strong>Paths</strong> to all of the dicom images (not as straightforward as you'd think)</li>\n<li>Bounding Boxes Melted Into <strong>Individual Rows</strong></li>\n<li><strong>Bounding Box Locations As Columns</strong> (xmin, ymin, xmax, ymax)</li>\n<li><strong>Fractional Bounding Box Coordinates</strong></li>\n<li>etc.</li>\n</ul>\n<p>Here is the full list of columns:</p>\n<ul>\n<li>id</li>\n<li>dcm_path</li>\n<li>xmin</li>\n<li>ymin</li>\n<li>xmax</li>\n<li>ymax</li>\n<li>human_label</li>\n<li>negative</li>\n<li>typical</li>\n<li>indeterminate</li>\n<li>atypical</li>\n<li>StudyInstanceUID</li>\n<li>SeriesInstanceUID</li>\n<li>ImageInstanceUID</li>\n<li>img_shape</li>\n<li>width</li>\n<li>height</li>\n<li>frac_xmin</li>\n<li>frac_xmax</li>\n<li>frac_ymin</li>\n<li>frac_ymax</li>\n<li>integer_label</li>\n</ul>",
  "messages": [
    {
      "id": "1314057",
      "postDate": "05/18/2021 23:21:23",
      "content": "<p>Hi there.</p>\n<p>I am working on an <a href=\"https://www.kaggle.com/dschettler8845/visual-in-depth-eda-siim-covid19-detection/output\" target=\"_blank\">EDA</a>, see here, and one of the outcomes is that I ended up with a data-frame/csv containing lots of useful information. I created a <a href=\"https://www.kaggle.com/dschettler8845/siim-covid19-updated-train-labels\" target=\"_blank\">dataset</a> for it to allow for easy access.</p>\n<p>I'm hoping this will save some people some time.</p>\n<p>Some examples:</p>\n<ul>\n<li><strong>Merged the two provided csvs into one on StudyInstanceUID</strong></li>\n<li><strong>Image Shape</strong> (takes a while to compute but now you don't have to)</li>\n<li><strong>Paths</strong> to all of the dicom images (not as straightforward as you'd think)</li>\n<li>Bounding Boxes Melted Into <strong>Individual Rows</strong></li>\n<li><strong>Bounding Box Locations As Columns</strong> (xmin, ymin, xmax, ymax)</li>\n<li><strong>Fractional Bounding Box Coordinates</strong></li>\n<li>etc.</li>\n</ul>\n<p>Here is the full list of columns:</p>\n<ul>\n<li>id</li>\n<li>dcm_path</li>\n<li>xmin</li>\n<li>ymin</li>\n<li>xmax</li>\n<li>ymax</li>\n<li>human_label</li>\n<li>negative</li>\n<li>typical</li>\n<li>indeterminate</li>\n<li>atypical</li>\n<li>StudyInstanceUID</li>\n<li>SeriesInstanceUID</li>\n<li>ImageInstanceUID</li>\n<li>img_shape</li>\n<li>width</li>\n<li>height</li>\n<li>frac_xmin</li>\n<li>frac_xmax</li>\n<li>frac_ymin</li>\n<li>frac_ymax</li>\n<li>integer_label</li>\n</ul>",
      "rawMarkdown": "Hi there.\n\nI am working on an [EDA](https://www.kaggle.com/dschettler8845/visual-in-depth-eda-siim-covid19-detection/output), see here, and one of the outcomes is that I ended up with a data-frame/csv containing lots of useful information. I created a [dataset](https://www.kaggle.com/dschettler8845/siim-covid19-updated-train-labels) for it to allow for easy access.\n\nI'm hoping this will save some people some time.\n\nSome examples:\n  - **Merged the two provided csvs into one on StudyInstanceUID**\n  - **Image Shape** (takes a while to compute but now you don't have to)\n  - **Paths** to all of the dicom images (not as straightforward as you'd think)\n  - Bounding Boxes Melted Into **Individual Rows**\n  - **Bounding Box Locations As Columns** (xmin, ymin, xmax, ymax)\n  - **Fractional Bounding Box Coordinates**\n  - etc.\n\nHere is the full list of columns:\n* id\n* dcm_path\n* xmin\n* ymin\n* xmax\n* ymax\n* human_label\n* negative\n* typical\n* indeterminate\n* atypical\n* StudyInstanceUID\n* SeriesInstanceUID\n* ImageInstanceUID\n* img_shape\n* width\n* height\n* frac_xmin\n* frac_xmax\n* frac_ymin\n* frac_ymax\n* integer_label",
      "votes": null
    },
    {
      "id": "1320953",
      "postDate": "05/24/2021 12:46:12",
      "content": "<p>the x y in the original data csv are center_x and center_y or something else ? The person who posted the competition should have told at least this</p>",
      "rawMarkdown": "the x y in the original data csv are center_x and center_y or something else ? The person who posted the competition should have told at least this",
      "votes": null
    },
    {
      "id": "1321500",
      "postDate": "05/24/2021 18:01:59",
      "content": "<p>Depending on the column the xy coordinates are either (xmin, ymin, width, height) or (xmin, ymin, xmax, ymax).</p>\n<p>I prefer the later so I used that format.</p>",
      "rawMarkdown": "Depending on the column the xy coordinates are either (xmin, ymin, width, height) or (xmin, ymin, xmax, ymax).\n\nI prefer the later so I used that format.",
      "votes": null
    },
    {
      "id": "1396455",
      "postDate": "07/22/2021 06:22:14",
      "content": "<p>Hi Darien,<br>\nWhat do you mean by \"depending on the column\" in your previous comment?</p>",
      "rawMarkdown": "Hi Darien,\nWhat do you mean by \"depending on the column\" in your previous comment?",
      "votes": null
    },
    {
      "id": "1396751",
      "postDate": "07/22/2021 13:14:54",
      "content": "<p>There is one column that reports data using the xmin,ymin,width,height format and one column that reports it as xmin,ymin,xmax,ymax. Or that was my understanding anyway.</p>",
      "rawMarkdown": "There is one column that reports data using the xmin,ymin,width,height format and one column that reports it as xmin,ymin,xmax,ymax. Or that was my understanding anyway.",
      "votes": null
    },
    {
      "id": "1397579",
      "postDate": "07/23/2021 10:06:38",
      "content": "<p>Got it.. Thanks</p>",
      "rawMarkdown": "Got it.. Thanks",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1320953,
      "author_name": "abhiagwl",
      "author_url": "",
      "post_date": "05/24/2021 12:46:12",
      "content": "<p>the x y in the original data csv are center_x and center_y or something else ? The person who posted the competition should have told at least this</p>",
      "votes": null,
      "replies": [
        {
          "id": 1321500,
          "author_name": "dschettler8845",
          "author_url": "",
          "post_date": "05/24/2021 18:01:59",
          "content": "<p>Depending on the column the xy coordinates are either (xmin, ymin, width, height) or (xmin, ymin, xmax, ymax).</p>\n<p>I prefer the later so I used that format.</p>",
          "votes": null,
          "replies": []
        }
      ]
    },
    {
      "id": 1396455,
      "author_name": "abhisekdash37",
      "author_url": "",
      "post_date": "07/22/2021 06:22:14",
      "content": "<p>Hi Darien,<br>\nWhat do you mean by \"depending on the column\" in your previous comment?</p>",
      "votes": null,
      "replies": [
        {
          "id": 1396751,
          "author_name": "dschettler8845",
          "author_url": "",
          "post_date": "07/22/2021 13:14:54",
          "content": "<p>There is one column that reports data using the xmin,ymin,width,height format and one column that reports it as xmin,ymin,xmax,ymax. Or that was my understanding anyway.</p>",
          "votes": null,
          "replies": []
        },
        {
          "id": 1397579,
          "author_name": "abhisekdash37",
          "author_url": "",
          "post_date": "07/23/2021 10:06:38",
          "content": "<p>Got it.. Thanks</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1314057": "Hi there.\n\nI am working on an [EDA](https://www.kaggle.com/dschettler8845/visual-in-depth-eda-siim-covid19-detection/output), see here, and one of the outcomes is that I ended up with a data-frame/csv containing lots of useful information. I created a [dataset](https://www.kaggle.com/dschettler8845/siim-covid19-updated-train-labels) for it to allow for easy access.\n\nI'm hoping this will save some people some time.\n\nSome examples:\n  - **Merged the two provided csvs into one on StudyInstanceUID**\n  - **Image Shape** (takes a while to compute but now you don't have to)\n  - **Paths** to all of the dicom images (not as straightforward as you'd think)\n  - Bounding Boxes Melted Into **Individual Rows**\n  - **Bounding Box Locations As Columns** (xmin, ymin, xmax, ymax)\n  - **Fractional Bounding Box Coordinates**\n  - etc.\n\nHere is the full list of columns:\n* id\n* dcm_path\n* xmin\n* ymin\n* xmax\n* ymax\n* human_label\n* negative\n* typical\n* indeterminate\n* atypical\n* StudyInstanceUID\n* SeriesInstanceUID\n* ImageInstanceUID\n* img_shape\n* width\n* height\n* frac_xmin\n* frac_xmax\n* frac_ymin\n* frac_ymax\n* integer_label",
    "1320953": "the x y in the original data csv are center_x and center_y or something else ? The person who posted the competition should have told at least this",
    "1321500": "Depending on the column the xy coordinates are either (xmin, ymin, width, height) or (xmin, ymin, xmax, ymax).\n\nI prefer the later so I used that format.",
    "1396455": "Hi Darien,\nWhat do you mean by \"depending on the column\" in your previous comment?",
    "1396751": "There is one column that reports data using the xmin,ymin,width,height format and one column that reports it as xmin,ymin,xmax,ymax. Or that was my understanding anyway.",
    "1397579": "Got it.. Thanks"
  },
  "source": "meta"
}