{
  "id": 340803,
  "title": "Mismatch in fractures area between train_bounding_boxes and segmentations",
  "url": "/competitions/rsna-2022-cervical-spine-fracture-detection/discussion/340803",
  "author_name": "Bùi Nhật Trường",
  "post_date": "2022-07-31T04:36:00.727000",
  "votes": 4,
  "comment_count": 4,
  "views": 0,
  "content": "<p>From the Data Description:</p>\n<p>train_bounding_boxes.csv Bounding boxes for a subset of the fractures in the training set.<br>\nsegmentations/ Pixel level annotations for a subset of the fractures in the training set.</p>\n<p>StudyInstanceUID checked: 1.2.826.0.1.3680043.11827<br>\nSlice: 76</p>\n<p>Segmentation image:<br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2F9540138001560bd98aea8f86c36fe456%2FSegm.PNG?generation=1659241952267989&amp;alt=media\" alt=\"\"></p>\n<p>Dcm image:<br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2F852b934854177bbc2a3059ddb64ef79a%2FDcm.PNG?generation=1659241974465266&amp;alt=media\" alt=\"\"></p>\n<p>Draw bbox from train_bounding_boxes.csv image (black rectangle box):<br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2Fbd09e895032f2c5bae67ebfb40c79dd7%2FBbox.PNG?generation=1659242037460621&amp;alt=media\" alt=\"\"></p>\n<p>Anyone get the same result ? Or am i missing/doing something wrong ?</p>",
  "messages": [
    {
      "id": 1877909,
      "postDate": "2022-07-31T04:36:00.727Z",
      "content": "<p>From the Data Description:</p>\n<p>train_bounding_boxes.csv Bounding boxes for a subset of the fractures in the training set.<br>\nsegmentations/ Pixel level annotations for a subset of the fractures in the training set.</p>\n<p>StudyInstanceUID checked: 1.2.826.0.1.3680043.11827<br>\nSlice: 76</p>\n<p>Segmentation image:<br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2F9540138001560bd98aea8f86c36fe456%2FSegm.PNG?generation=1659241952267989&amp;alt=media\" alt=\"\"></p>\n<p>Dcm image:<br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2F852b934854177bbc2a3059ddb64ef79a%2FDcm.PNG?generation=1659241974465266&amp;alt=media\" alt=\"\"></p>\n<p>Draw bbox from train_bounding_boxes.csv image (black rectangle box):<br>\n<img src=\"https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2Fbd09e895032f2c5bae67ebfb40c79dd7%2FBbox.PNG?generation=1659242037460621&amp;alt=media\" alt=\"\"></p>\n<p>Anyone get the same result ? Or am i missing/doing something wrong ?</p>",
      "rawMarkdown": "From the Data Description:\n\ntrain_bounding_boxes.csv Bounding boxes for a subset of the fractures in the training set.\nsegmentations/ Pixel level annotations for a subset of the fractures in the training set.\n\nStudyInstanceUID checked: 1.2.826.0.1.3680043.11827\nSlice: 76\n\nSegmentation image:\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2F9540138001560bd98aea8f86c36fe456%2FSegm.PNG?generation=1659241952267989&alt=media)\n\nDcm image:\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2F852b934854177bbc2a3059ddb64ef79a%2FDcm.PNG?generation=1659241974465266&alt=media)\n\nDraw bbox from train_bounding_boxes.csv image (black rectangle box):\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2Fbd09e895032f2c5bae67ebfb40c79dd7%2FBbox.PNG?generation=1659242037460621&alt=media)\n\nAnyone get the same result ? Or am i missing/doing something wrong ?",
      "votes": 4
    },
    {
      "id": 1878006,
      "postDate": "2022-07-31T05:54:51.703Z",
      "content": "<p>I do not see anything wrong?<br>\nI will have to correct you a bit, to my understanding, mask is the bone mask, and weather the bone is fractured or not has nothing to do with mask, but the box indicates the area where the bone is fractured… so box and mask represent two different things.</p>",
      "rawMarkdown": "I do not see anything wrong?\nI will have to correct you a bit, to my understanding, mask is the bone mask, and weather the bone is fractured or not has nothing to do with mask, but the box indicates the area where the bone is fractured... so box and mask represent two different things.",
      "votes": 2,
      "replies": [
        {
          "id": 1878048,
          "postDate": "2022-07-31T06:50:28.690Z",
          "content": "<p>If u put in that way, then it totally make sense. But Data description said: \"Segmentations/ Pixel level annotations for a subset of the <strong>fractures</strong> in the training set\". So i am still unsure about this. <a href=\"https://www.kaggle.com/felipekitamura\" target=\"_blank\">@felipekitamura</a> Can u confirm this ?</p>",
          "rawMarkdown": "If u put in that way, then it totally make sense. But Data description said: \"Segmentations/ Pixel level annotations for a subset of the **fractures** in the training set\". So i am still unsure about this. @felipekitamura Can u confirm this ?",
          "votes": 2
        },
        {
          "id": 1881747,
          "postDate": "2022-08-02T18:29:14.377Z",
          "content": "<p>Thanks for mentioning that. The sentence should read like \"Segmentations/ Pixel level annotations for a subset of the  training set\". We will correct it.</p>",
          "rawMarkdown": "Thanks for mentioning that. The sentence should read like \"Segmentations/ Pixel level annotations for a subset of the ~~fractures in the~~ training set\". We will correct it.",
          "votes": 1
        },
        {
          "id": 1883609,
          "postDate": "2022-08-04T00:51:14.223Z",
          "content": "<p>Ok, thank you for your answer</p>",
          "rawMarkdown": "Ok, thank you for your answer"
        }
      ]
    }
  ],
  "comments": [
    {
      "id": 1878006,
      "author_name": "Harshit Sheoran",
      "author_url": "",
      "post_date": "2022-07-31T05:54:51.703000",
      "content": "<p>I do not see anything wrong?<br>\nI will have to correct you a bit, to my understanding, mask is the bone mask, and weather the bone is fractured or not has nothing to do with mask, but the box indicates the area where the bone is fractured… so box and mask represent two different things.</p>",
      "votes": 2,
      "replies": [
        {
          "id": 1878048,
          "author_name": "Bùi Nhật Trường",
          "author_url": "",
          "post_date": "2022-07-31T06:50:28.690000",
          "content": "<p>If u put in that way, then it totally make sense. But Data description said: \"Segmentations/ Pixel level annotations for a subset of the <strong>fractures</strong> in the training set\". So i am still unsure about this. <a href=\"https://www.kaggle.com/felipekitamura\" target=\"_blank\">@felipekitamura</a> Can u confirm this ?</p>",
          "votes": 2,
          "replies": []
        },
        {
          "id": 1881747,
          "author_name": "FelipeKitamura, MD, PhD",
          "author_url": "",
          "post_date": "2022-08-02T18:29:14.377000",
          "content": "<p>Thanks for mentioning that. The sentence should read like \"Segmentations/ Pixel level annotations for a subset of the  training set\". We will correct it.</p>",
          "votes": 1,
          "replies": []
        },
        {
          "id": 1883609,
          "author_name": "Bùi Nhật Trường",
          "author_url": "",
          "post_date": "2022-08-04T00:51:14.223000",
          "content": "<p>Ok, thank you for your answer</p>",
          "votes": 0,
          "replies": []
        }
      ]
    }
  ],
  "raw_markdown_by_id": {
    "1877909": "From the Data Description:\n\ntrain_bounding_boxes.csv Bounding boxes for a subset of the fractures in the training set.\nsegmentations/ Pixel level annotations for a subset of the fractures in the training set.\n\nStudyInstanceUID checked: 1.2.826.0.1.3680043.11827\nSlice: 76\n\nSegmentation image:\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2F9540138001560bd98aea8f86c36fe456%2FSegm.PNG?generation=1659241952267989&alt=media)\n\nDcm image:\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2F852b934854177bbc2a3059ddb64ef79a%2FDcm.PNG?generation=1659241974465266&alt=media)\n\nDraw bbox from train_bounding_boxes.csv image (black rectangle box):\n![](https://www.googleapis.com/download/storage/v1/b/kaggle-forum-message-attachments/o/inbox%2F6329919%2Fbd09e895032f2c5bae67ebfb40c79dd7%2FBbox.PNG?generation=1659242037460621&alt=media)\n\nAnyone get the same result ? Or am i missing/doing something wrong ?",
    "1878006": "I do not see anything wrong?\nI will have to correct you a bit, to my understanding, mask is the bone mask, and weather the bone is fractured or not has nothing to do with mask, but the box indicates the area where the bone is fractured... so box and mask represent two different things."
  }
}