{
  "id": 174329,
  "title": "contextual information - sorry to bother again",
  "url": "/competitions/siim-isic-melanoma-classification/discussion/174329",
  "author_name": "",
  "post_date": "2020-08-13T07:23:28.518492100Z",
  "votes": null,
  "comment_count": 2,
  "views": 0,
  "content": "<p>\"Patient-level contextual information\" is any information gleaned from a patient's overall set of images, versus information from a single image. Age and sex are not outlined as patient-level contextual information for the purposes of special prize eligibility.</p>\n<p><a href=\"https://www.kaggle.com/juliaelliott\" target=\"_blank\">@juliaelliott</a> I had the opinion that sex and age are the only features that could be used for feature-based predictions, based on sentence above ( as they are outlined). But after rethinking I don't see that the other features (in train.csv), as well as derived features as mean color, are different in nature when used without patient_id information. Can you please clarify which reading is correct?</p>\n<p>Thx and best regards<br>\nRoman</p>",
  "messages": [
    {
      "id": "968685",
      "postDate": "08/13/2020 07:23:28",
      "content": "<p>\"Patient-level contextual information\" is any information gleaned from a patient's overall set of images, versus information from a single image. Age and sex are not outlined as patient-level contextual information for the purposes of special prize eligibility.</p>\n<p><a href=\"https://www.kaggle.com/juliaelliott\" target=\"_blank\">@juliaelliott</a> I had the opinion that sex and age are the only features that could be used for feature-based predictions, based on sentence above ( as they are outlined). But after rethinking I don't see that the other features (in train.csv), as well as derived features as mean color, are different in nature when used without patient_id information. Can you please clarify which reading is correct?</p>\n<p>Thx and best regards<br>\nRoman</p>",
      "rawMarkdown": "\"Patient-level contextual information\" is any information gleaned from a patient's overall set of images, versus information from a single image. Age and sex are not outlined as patient-level contextual information for the purposes of special prize eligibility.\n\n@juliaelliott I had the opinion that sex and age are the only features that could be used for feature-based predictions, based on sentence above ( as they are outlined). But after rethinking I don't see that the other features (in train.csv), as well as derived features as mean color, are different in nature when used without patient_id information. Can you please clarify which reading is correct?\n\nThx and best regards\nRoman",
      "votes": null
    },
    {
      "id": "969565",
      "postDate": "08/13/2020 19:14:16",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/romanweilguny\" target=\"_blank\">@romanweilguny</a> -- appreciate the question. What this is intended to specify is that in order to be considered \"with context,\" you'll have to make meaningful use of patient _id that considers a patient's overall set of images, and not simply age and sex or other features that are not truly aggregating context across a patient's images. Does that help?</p>",
      "rawMarkdown": "Hi @romanweilguny -- appreciate the question. What this is intended to specify is that in order to be considered \"with context,\" you'll have to make meaningful use of patient _id that considers a patient's overall set of images, and not simply age and sex or other features that are not truly aggregating context across a patient's images. Does that help?",
      "votes": null
    },
    {
      "id": "969598",
      "postDate": "08/13/2020 19:42:36",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/juliaelliott\" target=\"_blank\">@juliaelliott</a> -- So this was probably a misunderstanding due to my English and eg. anatom_site is allowed as a feature.<br>\nthx for clearing this.</p>",
      "rawMarkdown": "Hi @juliaelliott -- So this was probably a misunderstanding due to my English and eg. anatom_site is allowed as a feature.\nthx for clearing this.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 969565,
      "author_name": "juliaelliott",
      "author_url": "",
      "post_date": "08/13/2020 19:14:16",
      "content": "<p>Hi <a href=\"https://www.kaggle.com/romanweilguny\" target=\"_blank\">@romanweilguny</a> -- appreciate the question. What this is intended to specify is that in order to be considered \"with context,\" you'll have to make meaningful use of patient _id that considers a patient's overall set of images, and not simply age and sex or other features that are not truly aggregating context across a patient's images. Does that help?</p>",
      "votes": null,
      "replies": [
        {
          "id": 969598,
          "author_name": "romanweilguny",
          "author_url": "",
          "post_date": "08/13/2020 19:42:36",
          "content": "<p>Hi <a href=\"https://www.kaggle.com/juliaelliott\" target=\"_blank\">@juliaelliott</a> -- So this was probably a misunderstanding due to my English and eg. anatom_site is allowed as a feature.<br>\nthx for clearing this.</p>",
          "votes": null,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "968685": "\"Patient-level contextual information\" is any information gleaned from a patient's overall set of images, versus information from a single image. Age and sex are not outlined as patient-level contextual information for the purposes of special prize eligibility.\n\n@juliaelliott I had the opinion that sex and age are the only features that could be used for feature-based predictions, based on sentence above ( as they are outlined). But after rethinking I don't see that the other features (in train.csv), as well as derived features as mean color, are different in nature when used without patient_id information. Can you please clarify which reading is correct?\n\nThx and best regards\nRoman",
    "969565": "Hi @romanweilguny -- appreciate the question. What this is intended to specify is that in order to be considered \"with context,\" you'll have to make meaningful use of patient _id that considers a patient's overall set of images, and not simply age and sex or other features that are not truly aggregating context across a patient's images. Does that help?",
    "969598": "Hi @juliaelliott -- So this was probably a misunderstanding due to my English and eg. anatom_site is allowed as a feature.\nthx for clearing this."
  },
  "source": "meta"
}