{
  "id": 216732,
  "title": "Question About Training Image Level Labels",
  "url": "/competitions/hpa-single-cell-image-classification/discussion/216732",
  "author_name": "Darien Schettler",
  "post_date": "2021-02-03T20:09:15",
  "votes": 3,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Hi there.</p>\n<p>I had a question regarding the training data image-level labels.</p>\n<p><br></p>\n<p><strong>What I Already Know (sanity check)</strong></p>\n<hr>\n<p>We know that if an image is labelled as <strong>Nucleoplasm</strong> that one, some, or all of the cells in the image have the protein of interest localized within the <strong>Nucleoplasm</strong>.</p>\n<hr>\n<p><br></p>\n<p><strong>What I Want To Know</strong></p>\n<hr>\n<p>Is it possible that in our image labelled ONLY as <strong>Nucleoplasm</strong> that the protein of interest may also be localized in other organelles/structures? i.e. is it possible that the <strong>Cytoplasm</strong> in a couple of cells may contain some protein of interest but it was not called out at an image-level? If this is the case, what would the speculative impact of this be? Does it occur frequently? Or enough that it might have an impact?</p>\n<hr>\n<p><br></p>\n<p>Thanks for reading and thanks in advance for the insights!</p>",
  "messages": [
    {
      "id": 1187708,
      "postDate": "2021-02-05T16:43:26.127Z",
      "content": "<p>Hi,<br>\nYes there may be images where a single cell have a label that was neglected at the image level (for example a weak cytosolic pattern). Managing such weak-labels is part of this challenge, and it's impact is something we will analyze at the end of the challenge.</p>",
      "rawMarkdown": "Hi,\nYes there may be images where a single cell have a label that was neglected at the image level (for example a weak cytosolic pattern). Managing such weak-labels is part of this challenge, and it's impact is something we will analyze at the end of the challenge.",
      "votes": 6,
      "replies": [
        {
          "id": 1187715,
          "postDate": "2021-02-05T16:48:42.543Z",
          "content": "<p>Thanks Emma. I appreciate the response!</p>",
          "rawMarkdown": "Thanks Emma. I appreciate the response!"
        }
      ]
    },
    {
      "id": 1184870,
      "postDate": "2021-02-03T20:09:15Z",
      "content": "<p>Hi there.</p>\n<p>I had a question regarding the training data image-level labels.</p>\n<p><br></p>\n<p><strong>What I Already Know (sanity check)</strong></p>\n<hr>\n<p>We know that if an image is labelled as <strong>Nucleoplasm</strong> that one, some, or all of the cells in the image have the protein of interest localized within the <strong>Nucleoplasm</strong>.</p>\n<hr>\n<p><br></p>\n<p><strong>What I Want To Know</strong></p>\n<hr>\n<p>Is it possible that in our image labelled ONLY as <strong>Nucleoplasm</strong> that the protein of interest may also be localized in other organelles/structures? i.e. is it possible that the <strong>Cytoplasm</strong> in a couple of cells may contain some protein of interest but it was not called out at an image-level? If this is the case, what would the speculative impact of this be? Does it occur frequently? Or enough that it might have an impact?</p>\n<hr>\n<p><br></p>\n<p>Thanks for reading and thanks in advance for the insights!</p>",
      "rawMarkdown": "Hi there.\n\nI had a question regarding the training data image-level labels.\n\n<br>\n\n**What I Already Know (sanity check)**\n\n---\n\nWe know that if an image is labelled as **Nucleoplasm** that one, some, or all of the cells in the image have the protein of interest localized within the **Nucleoplasm**.\n\n---\n\n<br>\n\n**What I Want To Know**\n\n---\n\nIs it possible that in our image labelled ONLY as **Nucleoplasm** that the protein of interest may also be localized in other organelles/structures? i.e. is it possible that the **Cytoplasm** in a couple of cells may contain some protein of interest but it was not called out at an image-level? If this is the case, what would the speculative impact of this be? Does it occur frequently? Or enough that it might have an impact?\n\n---\n\n<br>\n\nThanks for reading and thanks in advance for the insights!",
      "votes": 3
    }
  ],
  "comments": [
    {
      "id": 1187708,
      "author_name": "Emma Lundberg",
      "author_url": "",
      "post_date": "2021-02-05T16:43:26.127000",
      "content": "<p>Hi,<br>\nYes there may be images where a single cell have a label that was neglected at the image level (for example a weak cytosolic pattern). Managing such weak-labels is part of this challenge, and it's impact is something we will analyze at the end of the challenge.</p>",
      "votes": 6,
      "replies": [
        {
          "id": 1187715,
          "author_name": "Darien Schettler",
          "author_url": "",
          "post_date": "2021-02-05T16:48:42.543000",
          "content": "<p>Thanks Emma. I appreciate the response!</p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1187708": "Hi,\nYes there may be images where a single cell have a label that was neglected at the image level (for example a weak cytosolic pattern). Managing such weak-labels is part of this challenge, and it's impact is something we will analyze at the end of the challenge.",
    "1184870": "Hi there.\n\nI had a question regarding the training data image-level labels.\n\n<br>\n\n**What I Already Know (sanity check)**\n\n---\n\nWe know that if an image is labelled as **Nucleoplasm** that one, some, or all of the cells in the image have the protein of interest localized within the **Nucleoplasm**.\n\n---\n\n<br>\n\n**What I Want To Know**\n\n---\n\nIs it possible that in our image labelled ONLY as **Nucleoplasm** that the protein of interest may also be localized in other organelles/structures? i.e. is it possible that the **Cytoplasm** in a couple of cells may contain some protein of interest but it was not called out at an image-level? If this is the case, what would the speculative impact of this be? Does it occur frequently? Or enough that it might have an impact?\n\n---\n\n<br>\n\nThanks for reading and thanks in advance for the insights!"
  }
}