{
  "id": 379061,
  "title": "RSNA BCD 1024x512 Cropped + Windowing",
  "url": "/competitions/rsna-breast-cancer-detection/discussion/379061",
  "author_name": "Paul Bacher",
  "post_date": "2023-01-17T23:03:49.199000",
  "votes": 17,
  "comment_count": 3,
  "views": 0,
  "content": "<p>Hi everybody!</p>\n<ul>\n<li>I share with you a dataset that I made and you can use to train a model for the competition.<br>\n<a href=\"https://www.kaggle.com/datasets/paulbacher/rsna-bcd-1024x512-preprocessed\" target=\"_blank\">RSNA BCD 1024x512 Cropped + Windowing</a><br>\nEverything has been cropped and the windowing has been applied on each scan. Also, all breasts are turns to the  same side (right). They are in PNG.</li>\n<li>I also made a lighter one if you want to make quick trainings, but it might not give great performances.<br>\n<a href=\"https://www.kaggle.com/datasets/paulbacher/rsna-preprocessed-512x256\" target=\"_blank\">RSNA BCD 512x256 Cropped + Windowing</a></li>\n<li><a href=\"https://www.kaggle.com/code/paulbacher/custom-preprocessor-rsna-breast-cancer\" target=\"_blank\">Here </a>is the preprocessor I used, in case you want to reuse it during inference or modify some of my preprocessing steps. I would also be glad to have your feedbacks!</li>\n<li>I'll work on generating the same dataset but in 2048x1024 if someone is interested!</li>\n</ul>\n<p>Cheerzzz</p>",
  "messages": [
    {
      "id": 2104592,
      "postDate": "2023-01-17T23:03:49.200Z",
      "content": "<p>Hi everybody!</p>\n<ul>\n<li>I share with you a dataset that I made and you can use to train a model for the competition.<br>\n<a href=\"https://www.kaggle.com/datasets/paulbacher/rsna-bcd-1024x512-preprocessed\" target=\"_blank\">RSNA BCD 1024x512 Cropped + Windowing</a><br>\nEverything has been cropped and the windowing has been applied on each scan. Also, all breasts are turns to the  same side (right). They are in PNG.</li>\n<li>I also made a lighter one if you want to make quick trainings, but it might not give great performances.<br>\n<a href=\"https://www.kaggle.com/datasets/paulbacher/rsna-preprocessed-512x256\" target=\"_blank\">RSNA BCD 512x256 Cropped + Windowing</a></li>\n<li><a href=\"https://www.kaggle.com/code/paulbacher/custom-preprocessor-rsna-breast-cancer\" target=\"_blank\">Here </a>is the preprocessor I used, in case you want to reuse it during inference or modify some of my preprocessing steps. I would also be glad to have your feedbacks!</li>\n<li>I'll work on generating the same dataset but in 2048x1024 if someone is interested!</li>\n</ul>\n<p>Cheerzzz</p>",
      "rawMarkdown": "Hi everybody!\n- I share with you a dataset that I made and you can use to train a model for the competition.\n[RSNA BCD 1024x512 Cropped + Windowing](https://www.kaggle.com/datasets/paulbacher/rsna-bcd-1024x512-preprocessed)\nEverything has been cropped and the windowing has been applied on each scan. Also, all breasts are turns to the  same side (right). They are in PNG.\n- I also made a lighter one if you want to make quick trainings, but it might not give great performances.\n[RSNA BCD 512x256 Cropped + Windowing](https://www.kaggle.com/datasets/paulbacher/rsna-preprocessed-512x256)\n- [Here ](https://www.kaggle.com/code/paulbacher/custom-preprocessor-rsna-breast-cancer)is the preprocessor I used, in case you want to reuse it during inference or modify some of my preprocessing steps. I would also be glad to have your feedbacks!\n- I'll work on generating the same dataset but in 2048x1024 if someone is interested!\n\nCheerzzz",
      "votes": 17
    },
    {
      "id": 2105910,
      "postDate": "2023-01-18T20:24:56.013Z",
      "content": "<p>Nice work, did you use it? How was the score?</p>\n<p>The Devastator.</p>",
      "rawMarkdown": "Nice work, did you use it? How was the score?\n\nThe Devastator.\n",
      "votes": 1
    },
    {
      "id": 2109165,
      "postDate": "2023-01-21T07:25:10.360Z",
      "content": "<p>Data Engineers are heroes🎉</p>",
      "rawMarkdown": "Data Engineers are heroes🎉"
    },
    {
      "id": 2106612,
      "postDate": "2023-01-19T09:00:52.333Z",
      "content": "<p>waiting for your 2048*1024 dataset! it will help me a lot! thanks!</p>",
      "rawMarkdown": "waiting for your 2048*1024 dataset! it will help me a lot! thanks!"
    }
  ],
  "comments": [
    {
      "id": 2105910,
      "author_name": "The Devastator",
      "author_url": "",
      "post_date": "2023-01-18T20:24:56.013000",
      "content": "<p>Nice work, did you use it? How was the score?</p>\n<p>The Devastator.</p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 2109165,
      "author_name": "Brendan Leahey",
      "author_url": "",
      "post_date": "2023-01-21T07:25:10.360000",
      "content": "<p>Data Engineers are heroes🎉</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2106612,
      "author_name": "Ernnnn4u",
      "author_url": "",
      "post_date": "2023-01-19T09:00:52.333000",
      "content": "<p>waiting for your 2048*1024 dataset! it will help me a lot! thanks!</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2104592": "Hi everybody!\n- I share with you a dataset that I made and you can use to train a model for the competition.\n[RSNA BCD 1024x512 Cropped + Windowing](https://www.kaggle.com/datasets/paulbacher/rsna-bcd-1024x512-preprocessed)\nEverything has been cropped and the windowing has been applied on each scan. Also, all breasts are turns to the  same side (right). They are in PNG.\n- I also made a lighter one if you want to make quick trainings, but it might not give great performances.\n[RSNA BCD 512x256 Cropped + Windowing](https://www.kaggle.com/datasets/paulbacher/rsna-preprocessed-512x256)\n- [Here ](https://www.kaggle.com/code/paulbacher/custom-preprocessor-rsna-breast-cancer)is the preprocessor I used, in case you want to reuse it during inference or modify some of my preprocessing steps. I would also be glad to have your feedbacks!\n- I'll work on generating the same dataset but in 2048x1024 if someone is interested!\n\nCheerzzz",
    "2105910": "Nice work, did you use it? How was the score?\n\nThe Devastator.\n",
    "2109165": "Data Engineers are heroes🎉",
    "2106612": "waiting for your 2048*1024 dataset! it will help me a lot! thanks!"
  }
}