{
  "id": 170684,
  "title": "What is the relation between the SliceThickness and SliceLocation? Equal or not?",
  "url": "/competitions/osic-pulmonary-fibrosis-progression/discussion/170684",
  "author_name": "",
  "post_date": "2020-07-28T14:58:07.131737600Z",
  "votes": null,
  "comment_count": 1,
  "views": 0,
  "content": "<p>Logically, the <code>slice thickness</code> is the thickness of slice, which is related to the detector of CT machine. Sometimes, the <code>slice thickness</code> can be calculated by the  <code>slice_thickness = np.abs(slices[0].SliceLocation - slices[1].SliceLocation)</code>. However, it doesn't work for the patient, for instance with ID = ID00228637202259965313869. </p>\n\n<p>In the case with ID = ID00228637202259965313869, the <code>slice_thickness = 1</code>, but <code>np.abs(slices[0].SliceLocation - slices[1].SliceLocation) = 0.8</code></p>\n\n<p>This difference might cause a problem when I want to integrate the volume of chest. Which value can I use to show the thickness of chest?</p>",
  "messages": [
    {
      "id": "949320",
      "postDate": "07/28/2020 14:58:07",
      "content": "<p>Logically, the <code>slice thickness</code> is the thickness of slice, which is related to the detector of CT machine. Sometimes, the <code>slice thickness</code> can be calculated by the  <code>slice_thickness = np.abs(slices[0].SliceLocation - slices[1].SliceLocation)</code>. However, it doesn't work for the patient, for instance with ID = ID00228637202259965313869. </p>\n\n<p>In the case with ID = ID00228637202259965313869, the <code>slice_thickness = 1</code>, but <code>np.abs(slices[0].SliceLocation - slices[1].SliceLocation) = 0.8</code></p>\n\n<p>This difference might cause a problem when I want to integrate the volume of chest. Which value can I use to show the thickness of chest?</p>",
      "rawMarkdown": "Logically, the `slice thickness` is the thickness of slice, which is related to the detector of CT machine. Sometimes, the `slice thickness` can be calculated by the  `slice_thickness = np.abs(slices[0].SliceLocation - slices[1].SliceLocation) `. However, it doesn't work for the patient, for instance with ID = ID00228637202259965313869. \n\nIn the case with ID = ID00228637202259965313869, the `slice_thickness = 1`, but `np.abs(slices[0].SliceLocation - slices[1].SliceLocation) = 0.8`\n\nThis difference might cause a problem when I want to integrate the volume of chest. Which value can I use to show the thickness of chest?",
      "votes": null
    },
    {
      "id": "949523",
      "postDate": "07/28/2020 17:27:09",
      "content": "<p>My interpretation is that the <code>.SliceLocation</code> attribute is the value on the z-axis.</p>\n\n<p>You can probably derive them using the <code>.SliceThickness</code> attribute, but watch out, the slices are not necessarily in order. Personally I use the <code>.InstanceNumber</code> attribute as a vertical \"index\" of the image files (this attribute seems to match the number used as the file names, so you can probably also use that).</p>",
      "rawMarkdown": "My interpretation is that the `.SliceLocation` attribute is the value on the z-axis.\n\nYou can probably derive them using the `.SliceThickness` attribute, but watch out, the slices are not necessarily in order. Personally I use the `.InstanceNumber` attribute as a vertical \"index\" of the image files (this attribute seems to match the number used as the file names, so you can probably also use that).",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 949523,
      "author_name": "archimedus",
      "author_url": "",
      "post_date": "07/28/2020 17:27:09",
      "content": "<p>My interpretation is that the <code>.SliceLocation</code> attribute is the value on the z-axis.</p>\n\n<p>You can probably derive them using the <code>.SliceThickness</code> attribute, but watch out, the slices are not necessarily in order. Personally I use the <code>.InstanceNumber</code> attribute as a vertical \"index\" of the image files (this attribute seems to match the number used as the file names, so you can probably also use that).</p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "949320": "Logically, the `slice thickness` is the thickness of slice, which is related to the detector of CT machine. Sometimes, the `slice thickness` can be calculated by the  `slice_thickness = np.abs(slices[0].SliceLocation - slices[1].SliceLocation) `. However, it doesn't work for the patient, for instance with ID = ID00228637202259965313869. \n\nIn the case with ID = ID00228637202259965313869, the `slice_thickness = 1`, but `np.abs(slices[0].SliceLocation - slices[1].SliceLocation) = 0.8`\n\nThis difference might cause a problem when I want to integrate the volume of chest. Which value can I use to show the thickness of chest?",
    "949523": "My interpretation is that the `.SliceLocation` attribute is the value on the z-axis.\n\nYou can probably derive them using the `.SliceThickness` attribute, but watch out, the slices are not necessarily in order. Personally I use the `.InstanceNumber` attribute as a vertical \"index\" of the image files (this attribute seems to match the number used as the file names, so you can probably also use that)."
  },
  "source": "meta"
}