{
  "id": 18700,
  "title": "Left ventricle segmentation",
  "url": "/competitions/second-annual-data-science-bowl/discussion/18700",
  "author_name": "",
  "post_date": "2016-02-02T11:40:13.187Z",
  "votes": null,
  "comment_count": 4,
  "views": 1212,
  "content": "<p>Some images are hard to segment. Since I'm not an expert, can someone explain where is the LV in those images (mainly the images 1 to 5) and how an expert would segment them.</p>\n\n<p>Thanks</p>\n\n<p><img src=\"http://lv.png/\" alt=\"enter image description here\" title></p>",
  "messages": [
    {
      "id": "106599",
      "postDate": "02/02/2016 11:40:13",
      "content": "<p>Some images are hard to segment. Since I'm not an expert, can someone explain where is the LV in those images (mainly the images 1 to 5) and how an expert would segment them.</p>\n\n<p>Thanks</p>\n\n<p><img src=\"http://lv.png/\" alt=\"enter image description here\" title></p>",
      "rawMarkdown": "Some images are hard to segment. Since I'm not an expert, can someone explain where is the LV in those images (mainly the images 1 to 5) and how an expert would segment them.\r\n\r\nThanks\r\n\r\n![enter image description here][1]\r\n\r\n\r\n  [1]: http://LV.png",
      "votes": null
    },
    {
      "id": "106609",
      "postDate": "02/02/2016 12:53:41",
      "content": "<p>yes, specifically the following two sequences from the same subject (train/1) are good examples I believe. One of them looks good, the other one does not seem to have the whole view. </p>\n\n<p>Are we to conclude that we can infer the same systole/diastole volumes from each of these sequences separately?</p>",
      "rawMarkdown": "yes, specifically the following two sequences from the same subject (train/1) are good examples I believe. One of them looks good, the other one does not seem to have the whole view. \r\n\r\nAre we to conclude that we can infer the same systole/diastole volumes from each of these sequences separately?",
      "votes": null
    },
    {
      "id": "106725",
      "postDate": "02/03/2016 14:52:26",
      "content": "<p>I am surprised that nobody is addressing this question.  From the two sax recordings I posted above, would you even bother segmenting sax1_5 to compute the volume while you already have sax1_8 which looks more like the images in the tutorials and the video?</p>",
      "rawMarkdown": "I am surprised that nobody is addressing this question.  From the two sax recordings I posted above, would you even bother segmenting sax1_5 to compute the volume while you already have sax1_8 which looks more like the images in the tutorials and the video?",
      "votes": null
    },
    {
      "id": "106735",
      "postDate": "02/03/2016 15:33:24",
      "content": "<p>My understanding is for  given patients, the studies (e.g. sax1, sax2) are sets of slices at different physical locations.  The files within a study (e.g. sax1..0001.dcm, sax1..0005.dcm) represent changes in time and are invariant w.r.t. physical location.  I sketched up a representation of my understanding ... which could be misguided.</p>\n\n<p><img src=\"https://www.kaggle.com/blobs/download/forum-message-attachment-files/3620/heartscans.png\" alt=\"sketch\" title></p>\n\n<p>EDIT:  Oops, bottom text on the left should have read 'sax2 ... 0001.dcm'.</p>",
      "rawMarkdown": "My understanding is for  given patients, the studies (e.g. sax1, sax2) are sets of slices at different physical locations.  The files within a study (e.g. sax1..0001.dcm, sax1..0005.dcm) represent changes in time and are invariant w.r.t. physical location.  I sketched up a representation of my understanding ... which could be misguided.\r\n\r\n![sketch][1]\r\n\r\nEDIT:  Oops, bottom text on the left should have read 'sax2 ... 0001.dcm'.\r\n\r\n  [1]: https://www.kaggle.com/blobs/download/forum-message-attachment-files/3620/heartscans.png",
      "votes": null
    },
    {
      "id": "106757",
      "postDate": "02/03/2016 18:19:02",
      "content": "<p>@Scott Smith, thanks. I should have started from the Fourier tutorial</p>",
      "rawMarkdown": "Scott Smith, thanks. I should have started from the Fourier tutorial",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 106609,
      "author_name": "",
      "author_url": "",
      "post_date": "02/02/2016 12:53:41",
      "content": "<p>yes, specifically the following two sequences from the same subject (train/1) are good examples I believe. One of them looks good, the other one does not seem to have the whole view. </p>\n\n<p>Are we to conclude that we can infer the same systole/diastole volumes from each of these sequences separately?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 106725,
      "author_name": "",
      "author_url": "",
      "post_date": "02/03/2016 14:52:26",
      "content": "<p>I am surprised that nobody is addressing this question.  From the two sax recordings I posted above, would you even bother segmenting sax1_5 to compute the volume while you already have sax1_8 which looks more like the images in the tutorials and the video?</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 106735,
      "author_name": "scsmith2",
      "author_url": "",
      "post_date": "02/03/2016 15:33:24",
      "content": "<p>My understanding is for  given patients, the studies (e.g. sax1, sax2) are sets of slices at different physical locations.  The files within a study (e.g. sax1..0001.dcm, sax1..0005.dcm) represent changes in time and are invariant w.r.t. physical location.  I sketched up a representation of my understanding ... which could be misguided.</p>\n\n<p><img src=\"https://www.kaggle.com/blobs/download/forum-message-attachment-files/3620/heartscans.png\" alt=\"sketch\" title></p>\n\n<p>EDIT:  Oops, bottom text on the left should have read 'sax2 ... 0001.dcm'.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 106757,
      "author_name": "",
      "author_url": "",
      "post_date": "02/03/2016 18:19:02",
      "content": "<p>@Scott Smith, thanks. I should have started from the Fourier tutorial</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "106599": "Some images are hard to segment. Since I'm not an expert, can someone explain where is the LV in those images (mainly the images 1 to 5) and how an expert would segment them.\r\n\r\nThanks\r\n\r\n![enter image description here][1]\r\n\r\n\r\n  [1]: http://LV.png",
    "106609": "yes, specifically the following two sequences from the same subject (train/1) are good examples I believe. One of them looks good, the other one does not seem to have the whole view. \r\n\r\nAre we to conclude that we can infer the same systole/diastole volumes from each of these sequences separately?",
    "106725": "I am surprised that nobody is addressing this question.  From the two sax recordings I posted above, would you even bother segmenting sax1_5 to compute the volume while you already have sax1_8 which looks more like the images in the tutorials and the video?",
    "106735": "My understanding is for  given patients, the studies (e.g. sax1, sax2) are sets of slices at different physical locations.  The files within a study (e.g. sax1..0001.dcm, sax1..0005.dcm) represent changes in time and are invariant w.r.t. physical location.  I sketched up a representation of my understanding ... which could be misguided.\r\n\r\n![sketch][1]\r\n\r\nEDIT:  Oops, bottom text on the left should have read 'sax2 ... 0001.dcm'.\r\n\r\n  [1]: https://www.kaggle.com/blobs/download/forum-message-attachment-files/3620/heartscans.png",
    "106757": "Scott Smith, thanks. I should have started from the Fourier tutorial"
  },
  "source": "meta"
}