{
  "id": 190197,
  "title": "Diff between -ve exam for PE and pe_present_on_image",
  "url": "/competitions/rsna-str-pulmonary-embolism-detection/discussion/190197",
  "author_name": "Jaideep",
  "post_date": "2020-10-10T15:37:17.373000",
  "votes": 1,
  "comment_count": 2,
  "views": 0,
  "content": "<p>Can some one exaplin the difference between the two. <br>\nI expect that if pe_present_on_image is 0 then -ve exam for pe should always be 1 which i very well see</p>\n<p>but below give non zero number,which i dont understand why…<br>\n<code>train_df[(train_df.negative_exam_for_pe==0) &amp; (train_df.pe_present_on_image==0) ].StudyInstanceUID.nunique()</code></p>",
  "messages": [
    {
      "id": 1045424,
      "postDate": "2020-10-10T16:11:27.183Z",
      "content": "<p>pe present on image is for the individual image</p>\n<p>All the rest of the labels are for the entire study</p>\n<p>So a study that is negative exam for PE will not have any images that are pe present on image</p>\n<p>If a study has any pe present in image, then it will not be exam negative for pe</p>",
      "rawMarkdown": "pe present on image is for the individual image\n\nAll the rest of the labels are for the entire study\n\nSo a study that is negative exam for PE will not have any images that are pe present on image\n\nIf a study has any pe present in image, then it will not be exam negative for pe",
      "votes": 1,
      "replies": [
        {
          "id": 1045451,
          "postDate": "2020-10-10T16:57:15.223Z",
          "content": "<p><a href=\"https://www.kaggle.com/richardepstein\" target=\"_blank\">@richardepstein</a>  thanks..<br>\nyes i just checked…Few more thing<br>\n1)   if -ve exam for pe is 1 which means entire study is  having no image with PE . <br>\nso can that person be assumed healthy ?</p>\n<p>2) What about the Study where some images have PE present as 0 , when i check the fields like left,right  central pe  have got label as 1 which i think are at study level. I m bit confused here  .<br>\nDoes it means that this study   if got 300 slices or images  then  for whichever image if field pe present on image is 1 then  the location of PE would be as per other study level fields like  left,right,central ,ratio fields ?</p>",
          "rawMarkdown": "@richardepstein  thanks..\nyes i just checked...Few more thing\n1)   if -ve exam for pe is 1 which means entire study is  having no image with PE . \nso can that person be assumed healthy ?\n\n\n2) What about the Study where some images have PE present as 0 , when i check the fields like left,right  central pe  have got label as 1 which i think are at study level. I m bit confused here  .\nDoes it means that this study   if got 300 slices or images  then  for whichever image if field pe present on image is 1 then  the location of PE would be as per other study level fields like  left,right,central ,ratio fields ?\n\n\n\n\n\n"
        }
      ]
    },
    {
      "id": 1045404,
      "postDate": "2020-10-10T15:37:17.373Z",
      "content": "<p>Can some one exaplin the difference between the two. <br>\nI expect that if pe_present_on_image is 0 then -ve exam for pe should always be 1 which i very well see</p>\n<p>but below give non zero number,which i dont understand why…<br>\n<code>train_df[(train_df.negative_exam_for_pe==0) &amp; (train_df.pe_present_on_image==0) ].StudyInstanceUID.nunique()</code></p>",
      "rawMarkdown": "Can some one exaplin the difference between the two. \nI expect that if pe_present_on_image is 0 then -ve exam for pe should always be 1 which i very well see\n\nbut below give non zero number,which i dont understand why...\n`train_df[(train_df.negative_exam_for_pe==0) & (train_df.pe_present_on_image==0) ].StudyInstanceUID.nunique()`",
      "votes": 1
    }
  ],
  "comments": [
    {
      "id": 1045424,
      "author_name": "quadcore/Richard Epstein",
      "author_url": "",
      "post_date": "2020-10-10T16:11:27.183000",
      "content": "<p>pe present on image is for the individual image</p>\n<p>All the rest of the labels are for the entire study</p>\n<p>So a study that is negative exam for PE will not have any images that are pe present on image</p>\n<p>If a study has any pe present in image, then it will not be exam negative for pe</p>",
      "votes": 1,
      "replies": [
        {
          "id": 1045451,
          "author_name": "Jaideep",
          "author_url": "",
          "post_date": "2020-10-10T16:57:15.223000",
          "content": "<p><a href=\"https://www.kaggle.com/richardepstein\" target=\"_blank\">@richardepstein</a>  thanks..<br>\nyes i just checked…Few more thing<br>\n1)   if -ve exam for pe is 1 which means entire study is  having no image with PE . <br>\nso can that person be assumed healthy ?</p>\n<p>2) What about the Study where some images have PE present as 0 , when i check the fields like left,right  central pe  have got label as 1 which i think are at study level. I m bit confused here  .<br>\nDoes it means that this study   if got 300 slices or images  then  for whichever image if field pe present on image is 1 then  the location of PE would be as per other study level fields like  left,right,central ,ratio fields ?</p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1045424": "pe present on image is for the individual image\n\nAll the rest of the labels are for the entire study\n\nSo a study that is negative exam for PE will not have any images that are pe present on image\n\nIf a study has any pe present in image, then it will not be exam negative for pe",
    "1045404": "Can some one exaplin the difference between the two. \nI expect that if pe_present_on_image is 0 then -ve exam for pe should always be 1 which i very well see\n\nbut below give non zero number,which i dont understand why...\n`train_df[(train_df.negative_exam_for_pe==0) & (train_df.pe_present_on_image==0) ].StudyInstanceUID.nunique()`"
  }
}