{
  "id": 370507,
  "title": "Site 2 difficult negative cases and biopsies",
  "url": "/competitions/rsna-breast-cancer-detection/discussion/370507",
  "author_name": "@kaggleqrdl",
  "post_date": "2022-12-04T23:36:19.169000",
  "votes": 2,
  "comment_count": 0,
  "views": 0,
  "content": "<p>I did some eda on all the tabular data, nothing complex, just groupby and some observations.  </p>\n<p><a href=\"https://www.kaggle.com/code/kaggleqrdl/non-image-eda\" target=\"_blank\">https://www.kaggle.com/code/kaggleqrdl/non-image-eda</a></p>\n<p>The data shows some statistical oddities for sure between the sites, but a big one I noticed was this:</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F9052057%2Fe564fb1178b05c69569952fa0a7f8575%2FScreenshot%202022-12-04%20162455.png?generation=1670196546577301&amp;alt=media\" alt=\"\"></p>\n<p>I would assume if a case was difficult, you'd be much more likely to get a biopsy.   In the above image it looks like 33% of all difficult cases get a biopsy at site 1 versus 4% at site 2.  I admit I found this surprising.</p>\n<p>Note that counts are 5007 difficult cases at site 1, versus 2691 at site 2.  </p>",
  "messages": [
    {
      "id": 2055251,
      "postDate": "2022-12-04T23:36:19.170Z",
      "content": "<p>I did some eda on all the tabular data, nothing complex, just groupby and some observations.  </p>\n<p><a href=\"https://www.kaggle.com/code/kaggleqrdl/non-image-eda\" target=\"_blank\">https://www.kaggle.com/code/kaggleqrdl/non-image-eda</a></p>\n<p>The data shows some statistical oddities for sure between the sites, but a big one I noticed was this:</p>\n<p><img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F9052057%2Fe564fb1178b05c69569952fa0a7f8575%2FScreenshot%202022-12-04%20162455.png?generation=1670196546577301&amp;alt=media\" alt=\"\"></p>\n<p>I would assume if a case was difficult, you'd be much more likely to get a biopsy.   In the above image it looks like 33% of all difficult cases get a biopsy at site 1 versus 4% at site 2.  I admit I found this surprising.</p>\n<p>Note that counts are 5007 difficult cases at site 1, versus 2691 at site 2.  </p>",
      "rawMarkdown": "I did some eda on all the tabular data, nothing complex, just groupby and some observations.  \n\nhttps://www.kaggle.com/code/kaggleqrdl/non-image-eda\n\nThe data shows some statistical oddities for sure between the sites, but a big one I noticed was this:\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F9052057%2Fe564fb1178b05c69569952fa0a7f8575%2FScreenshot%202022-12-04%20162455.png?generation=1670196546577301&alt=media)\n\nI would assume if a case was difficult, you'd be much more likely to get a biopsy.   In the above image it looks like 33% of all difficult cases get a biopsy at site 1 versus 4% at site 2.  I admit I found this surprising.\n\nNote that counts are 5007 difficult cases at site 1, versus 2691 at site 2.  \n\n\n",
      "votes": 2
    }
  ],
  "comments": [],
  "raw_markdown_by_id": {
    "2055251": "I did some eda on all the tabular data, nothing complex, just groupby and some observations.  \n\nhttps://www.kaggle.com/code/kaggleqrdl/non-image-eda\n\nThe data shows some statistical oddities for sure between the sites, but a big one I noticed was this:\n\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F9052057%2Fe564fb1178b05c69569952fa0a7f8575%2FScreenshot%202022-12-04%20162455.png?generation=1670196546577301&alt=media)\n\nI would assume if a case was difficult, you'd be much more likely to get a biopsy.   In the above image it looks like 33% of all difficult cases get a biopsy at site 1 versus 4% at site 2.  I admit I found this surprising.\n\nNote that counts are 5007 difficult cases at site 1, versus 2691 at site 2.  \n\n\n"
  }
}