{
  "id": 369583,
  "title": "Schrödinger Labels?",
  "url": "/competitions/rsna-breast-cancer-detection/discussion/369583",
  "author_name": "DeepUnderstanding",
  "post_date": "2022-11-30T15:20:25.297000",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>As I was going through the data I have seen some cases, where a unique patient is both marked 1 and 0 for Cancer label. Yes its for different images, but if the patient has cancer how can some of the labels  are marked 0.</p>\n<p>For example patient id = 32770, has cancer = 0 for the laterality = 'L' and cancer = 1 for 'R'.</p>\n<p>So are we supposed to look each image as a separate datapoint and not take the patient_id info for splitting the data?</p>",
  "messages": [
    {
      "id": 2050349,
      "postDate": "2022-11-30T17:20:56.763Z",
      "content": "<p>The cancer labels are per breast, not per patient.</p>",
      "rawMarkdown": "The cancer labels are per breast, not per patient.",
      "votes": 5,
      "replies": [
        {
          "id": 2050497,
          "postDate": "2022-11-30T19:50:59.007Z",
          "content": "<p>Oh I see it now, it means which breast of the patient has the cancer, rather than the patent having it, got it</p>",
          "rawMarkdown": "Oh I see it now, it means which breast of the patient has the cancer, rather than the patent having it, got it",
          "votes": 1
        }
      ]
    },
    {
      "id": 2050187,
      "postDate": "2022-11-30T15:20:25.297Z",
      "content": "<p>As I was going through the data I have seen some cases, where a unique patient is both marked 1 and 0 for Cancer label. Yes its for different images, but if the patient has cancer how can some of the labels  are marked 0.</p>\n<p>For example patient id = 32770, has cancer = 0 for the laterality = 'L' and cancer = 1 for 'R'.</p>\n<p>So are we supposed to look each image as a separate datapoint and not take the patient_id info for splitting the data?</p>",
      "rawMarkdown": "As I was going through the data I have seen some cases, where a unique patient is both marked 1 and 0 for Cancer label. Yes its for different images, but if the patient has cancer how can some of the labels  are marked 0.\n\nFor example patient id = 32770, has cancer = 0 for the laterality = 'L' and cancer = 1 for 'R'.\n\nSo are we supposed to look each image as a separate datapoint and not take the patient_id info for splitting the data?",
      "votes": 1
    }
  ],
  "comments": [
    {
      "id": 2050349,
      "author_name": "Sohier Dane",
      "author_url": "",
      "post_date": "2022-11-30T17:20:56.763000",
      "content": "<p>The cancer labels are per breast, not per patient.</p>",
      "votes": 5,
      "replies": [
        {
          "id": 2050497,
          "author_name": "DeepUnderstanding",
          "author_url": "",
          "post_date": "2022-11-30T19:50:59.007000",
          "content": "<p>Oh I see it now, it means which breast of the patient has the cancer, rather than the patent having it, got it</p>",
          "votes": 1,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "2050349": "The cancer labels are per breast, not per patient.",
    "2050187": "As I was going through the data I have seen some cases, where a unique patient is both marked 1 and 0 for Cancer label. Yes its for different images, but if the patient has cancer how can some of the labels  are marked 0.\n\nFor example patient id = 32770, has cancer = 0 for the laterality = 'L' and cancer = 1 for 'R'.\n\nSo are we supposed to look each image as a separate datapoint and not take the patient_id info for splitting the data?"
  }
}