{
  "id": 611002,
  "title": "Is the coordinates in train_localizers.csv the pixel positions?",
  "url": "/competitions/rsna-intracranial-aneurysm-detection/discussion/611002",
  "author_name": "Yuzhi Xu",
  "post_date": "2025-10-08T07:15:02.692000",
  "votes": 0,
  "comment_count": 2,
  "views": 0,
  "content": "<p>For example, in the first row of train_localizers.csv</p>\n<table>\n<thead>\n<tr>\n<th>SeriesInstanceUID</th>\n<th>SOPInstanceUID</th>\n<th>coordinates</th>\n<th>location</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>1.2.826.0.1.3680043.8.498.10005158603912009425635473100344077317</td>\n<td>1.2.826.0.1.3680043.8.498.10775329348174902199350466348663848346</td>\n<td>{'x': 258.3621186176837, 'y': 261.359900373599}</td>\n<td>Other Posterior Circulation</td>\n</tr>\n</tbody>\n</table>\n<p>Does it mean in the series directory <strong>\"1.2.826.0.1.3680043.8.498.10005158603912009425635473100344077317\"</strong> which is <strong>SeriesInstanceUID</strong> and the file <strong>1.2.826.0.1.3680043.8.498.10775329348174902199350466348663848346.dcm</strong> which is <strong>SOPInstanceUID</strong>, there is a Intracranial Aneurysm in <strong>row:258</strong>, <strong>col:261</strong> which are <strong>\"x\"</strong> and <strong>\"y\"</strong> roughly?</p>\n<p>I mean <strong>“x”</strong> and <strong>”y“</strong> are pixel coordinates, not world coordinates, right?</p>",
  "messages": [
    {
      "id": 3299474,
      "postDate": "2025-10-08T07:29:05.180Z",
      "content": "<p>If I'm right, yes, they're the direct slice pixel coordinates. No transforms needed.<br>\nBut no, y is row and x is column.</p>\n<p>EDIT : Slice[np.rint(y).astype(int),np.rint(x).astype(int)]</p>",
      "rawMarkdown": "If I'm right, yes, they're the direct slice pixel coordinates. No transforms needed.\nBut no, y is row and x is column.\n\nEDIT : Slice[np.rint(y).astype(int),np.rint(x).astype(int)]",
      "votes": 2,
      "replies": [
        {
          "id": 3299482,
          "postDate": "2025-10-08T08:08:34.600Z",
          "content": "<p>yes,  y is row and x is column. Thank you.</p>",
          "rawMarkdown": "yes,  y is row and x is column. Thank you."
        }
      ]
    },
    {
      "id": 3299473,
      "postDate": "2025-10-08T07:15:02.693Z",
      "content": "<p>For example, in the first row of train_localizers.csv</p>\n<table>\n<thead>\n<tr>\n<th>SeriesInstanceUID</th>\n<th>SOPInstanceUID</th>\n<th>coordinates</th>\n<th>location</th>\n</tr>\n</thead>\n<tbody>\n<tr>\n<td>1.2.826.0.1.3680043.8.498.10005158603912009425635473100344077317</td>\n<td>1.2.826.0.1.3680043.8.498.10775329348174902199350466348663848346</td>\n<td>{'x': 258.3621186176837, 'y': 261.359900373599}</td>\n<td>Other Posterior Circulation</td>\n</tr>\n</tbody>\n</table>\n<p>Does it mean in the series directory <strong>\"1.2.826.0.1.3680043.8.498.10005158603912009425635473100344077317\"</strong> which is <strong>SeriesInstanceUID</strong> and the file <strong>1.2.826.0.1.3680043.8.498.10775329348174902199350466348663848346.dcm</strong> which is <strong>SOPInstanceUID</strong>, there is a Intracranial Aneurysm in <strong>row:258</strong>, <strong>col:261</strong> which are <strong>\"x\"</strong> and <strong>\"y\"</strong> roughly?</p>\n<p>I mean <strong>“x”</strong> and <strong>”y“</strong> are pixel coordinates, not world coordinates, right?</p>",
      "rawMarkdown": "For example, in the first row of train_localizers.csv\n |SeriesInstanceUID  | SOPInstanceUID  |coordinates|location|\n| ---  | --- |\n|1.2.826.0.1.3680043.8.498.10005158603912009425635473100344077317 |  1.2.826.0.1.3680043.8.498.10775329348174902199350466348663848346|{'x': 258.3621186176837, 'y': 261.359900373599}|Other Posterior Circulation|\n\nDoes it mean in the series directory **\"1.2.826.0.1.3680043.8.498.10005158603912009425635473100344077317\"** which is **SeriesInstanceUID** and the file **1.2.826.0.1.3680043.8.498.10775329348174902199350466348663848346.dcm** which is **SOPInstanceUID**, there is a Intracranial Aneurysm in **row:258**, **col:261** which are **\"x\"** and **\"y\"** roughly?\n\nI mean **“x”** and **”y“** are pixel coordinates, not world coordinates, right?"
    }
  ],
  "comments": [
    {
      "id": 3299474,
      "author_name": "Ángel Jacinto Sánchez Ruiz",
      "author_url": "",
      "post_date": "2025-10-08T07:29:05.180000",
      "content": "<p>If I'm right, yes, they're the direct slice pixel coordinates. No transforms needed.<br>\nBut no, y is row and x is column.</p>\n<p>EDIT : Slice[np.rint(y).astype(int),np.rint(x).astype(int)]</p>",
      "votes": 2,
      "replies": [
        {
          "id": 3299482,
          "author_name": "Yuzhi Xu",
          "author_url": "",
          "post_date": "2025-10-08T08:08:34.600000",
          "content": "<p>yes,  y is row and x is column. Thank you.</p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "3299474": "If I'm right, yes, they're the direct slice pixel coordinates. No transforms needed.\nBut no, y is row and x is column.\n\nEDIT : Slice[np.rint(y).astype(int),np.rint(x).astype(int)]",
    "3299473": "For example, in the first row of train_localizers.csv\n |SeriesInstanceUID  | SOPInstanceUID  |coordinates|location|\n| ---  | --- |\n|1.2.826.0.1.3680043.8.498.10005158603912009425635473100344077317 |  1.2.826.0.1.3680043.8.498.10775329348174902199350466348663848346|{'x': 258.3621186176837, 'y': 261.359900373599}|Other Posterior Circulation|\n\nDoes it mean in the series directory **\"1.2.826.0.1.3680043.8.498.10005158603912009425635473100344077317\"** which is **SeriesInstanceUID** and the file **1.2.826.0.1.3680043.8.498.10775329348174902199350466348663848346.dcm** which is **SOPInstanceUID**, there is a Intracranial Aneurysm in **row:258**, **col:261** which are **\"x\"** and **\"y\"** roughly?\n\nI mean **“x”** and **”y“** are pixel coordinates, not world coordinates, right?"
  }
}