{
  "id": 245225,
  "title": "Papers on detection of covid 19  ",
  "url": "/competitions/siim-covid19-detection/discussion/245225",
  "author_name": "",
  "post_date": "2021-06-10T08:04:23.879232600Z",
  "votes": 1,
  "comment_count": 1,
  "views": 0,
  "content": "<p>No papers on detection of covid 19 via ML is suitable for clinical use </p>\n<p>Ref<br>\n<a href=\"https://twitter.com/haltakov/status/1402717293801508867?s=20\" target=\"_blank\">https://twitter.com/haltakov/status/1402717293801508867?s=20</a><br>\n<a href=\"https://www.radiologybusiness.com/topics/artificial-intelligence/imaging-ai-models-covid-19-diagnosis-clinical-use?utm_source=newsletter&amp;utm_medium=rb_news\" target=\"_blank\">https://www.radiologybusiness.com/topics/artificial-intelligence/imaging-ai-models-covid-19-diagnosis-clinical-use?utm_source=newsletter&amp;utm_medium=rb_news</a></p>\n<p>Nature paper <br>\n<a href=\"https://www.nature.com/articles/s42256-021-00307-0\" target=\"_blank\">https://www.nature.com/articles/s42256-021-00307-0</a></p>\n<p>2 papers of qure.ai that worked but were not included for study <br>\n<a href=\"https://www.nature.com/articles/s41598-020-79470-0\" target=\"_blank\">https://www.nature.com/articles/s41598-020-79470-0</a><br>\n<a href=\"https://link.springer.com/article/10.1007/s00330-020-07269-8\" target=\"_blank\">https://link.springer.com/article/10.1007/s00330-020-07269-8</a></p>",
  "messages": [
    {
      "id": "1343454",
      "postDate": "06/10/2021 08:04:23",
      "content": "<p>No papers on detection of covid 19 via ML is suitable for clinical use </p>\n<p>Ref<br>\n<a href=\"https://twitter.com/haltakov/status/1402717293801508867?s=20\" target=\"_blank\">https://twitter.com/haltakov/status/1402717293801508867?s=20</a><br>\n<a href=\"https://www.radiologybusiness.com/topics/artificial-intelligence/imaging-ai-models-covid-19-diagnosis-clinical-use?utm_source=newsletter&amp;utm_medium=rb_news\" target=\"_blank\">https://www.radiologybusiness.com/topics/artificial-intelligence/imaging-ai-models-covid-19-diagnosis-clinical-use?utm_source=newsletter&amp;utm_medium=rb_news</a></p>\n<p>Nature paper <br>\n<a href=\"https://www.nature.com/articles/s42256-021-00307-0\" target=\"_blank\">https://www.nature.com/articles/s42256-021-00307-0</a></p>\n<p>2 papers of qure.ai that worked but were not included for study <br>\n<a href=\"https://www.nature.com/articles/s41598-020-79470-0\" target=\"_blank\">https://www.nature.com/articles/s41598-020-79470-0</a><br>\n<a href=\"https://link.springer.com/article/10.1007/s00330-020-07269-8\" target=\"_blank\">https://link.springer.com/article/10.1007/s00330-020-07269-8</a></p>",
      "rawMarkdown": "No papers on detection of covid 19 via ML is suitable for clinical use \n\nRef\nhttps://twitter.com/haltakov/status/1402717293801508867?s=20\nhttps://www.radiologybusiness.com/topics/artificial-intelligence/imaging-ai-models-covid-19-diagnosis-clinical-use?utm_source=newsletter&utm_medium=rb_news\n\nNature paper \nhttps://www.nature.com/articles/s42256-021-00307-0\n\n2 papers of qure.ai that worked but were not included for study \nhttps://www.nature.com/articles/s41598-020-79470-0\nhttps://link.springer.com/article/10.1007/s00330-020-07269-8",
      "votes": null
    },
    {
      "id": "1343793",
      "postDate": "06/10/2021 12:50:30",
      "content": "<p>Thanks for sharing this.</p>\n<p>I write medical imaging software at my day job. I've spoken with several radiologist in-depth about this very subject. They have all stated that it is not possible to diagnose Covid-19 purely from a chest radiograph, nor a CT. The appearance of pneumonia along with patient history might lead a radiologist to suspect Covid-19, but he cannot discern Covid from other pneumonia/infiltrate findings. </p>\n<p>Of course some will say \"but what about ground-glass appearance on axial CT slices?\" I'll leave that for the radiologists to debate. In fact, I'd love to get some expert feedback here on this topic.</p>",
      "rawMarkdown": "Thanks for sharing this.\n\nI write medical imaging software at my day job. I've spoken with several radiologist in-depth about this very subject. They have all stated that it is not possible to diagnose Covid-19 purely from a chest radiograph, nor a CT. The appearance of pneumonia along with patient history might lead a radiologist to suspect Covid-19, but he cannot discern Covid from other pneumonia/infiltrate findings. \n\nOf course some will say \"but what about ground-glass appearance on axial CT slices?\" I'll leave that for the radiologists to debate. In fact, I'd love to get some expert feedback here on this topic.",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1343793,
      "author_name": "davidbroberts",
      "author_url": "",
      "post_date": "06/10/2021 12:50:30",
      "content": "<p>Thanks for sharing this.</p>\n<p>I write medical imaging software at my day job. I've spoken with several radiologist in-depth about this very subject. They have all stated that it is not possible to diagnose Covid-19 purely from a chest radiograph, nor a CT. The appearance of pneumonia along with patient history might lead a radiologist to suspect Covid-19, but he cannot discern Covid from other pneumonia/infiltrate findings. </p>\n<p>Of course some will say \"but what about ground-glass appearance on axial CT slices?\" I'll leave that for the radiologists to debate. In fact, I'd love to get some expert feedback here on this topic.</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1343454": "No papers on detection of covid 19 via ML is suitable for clinical use \n\nRef\nhttps://twitter.com/haltakov/status/1402717293801508867?s=20\nhttps://www.radiologybusiness.com/topics/artificial-intelligence/imaging-ai-models-covid-19-diagnosis-clinical-use?utm_source=newsletter&utm_medium=rb_news\n\nNature paper \nhttps://www.nature.com/articles/s42256-021-00307-0\n\n2 papers of qure.ai that worked but were not included for study \nhttps://www.nature.com/articles/s41598-020-79470-0\nhttps://link.springer.com/article/10.1007/s00330-020-07269-8",
    "1343793": "Thanks for sharing this.\n\nI write medical imaging software at my day job. I've spoken with several radiologist in-depth about this very subject. They have all stated that it is not possible to diagnose Covid-19 purely from a chest radiograph, nor a CT. The appearance of pneumonia along with patient history might lead a radiologist to suspect Covid-19, but he cannot discern Covid from other pneumonia/infiltrate findings. \n\nOf course some will say \"but what about ground-glass appearance on axial CT slices?\" I'll leave that for the radiologists to debate. In fact, I'd love to get some expert feedback here on this topic."
  },
  "source": "meta"
}