{
  "id": 197609,
  "title": "[PAPER | STARTER]: Segmentation of Glomeruli Within Trichrome Images Using Deep Learning",
  "url": "/competitions/hubmap-kidney-segmentation/discussion/197609",
  "author_name": "",
  "post_date": "2020-11-17T09:20:32.183599300Z",
  "votes": 5,
  "comment_count": 2,
  "views": 0,
  "content": "<p>In case, someone missed it, the paper linked under the data section of this competition actually provides a good summary on how to start with this competition. </p>\n<p>You can reference it <a href=\"https://www.kireports.org/article/S2468-0249%2819%2930155-X/abstract\" target=\"_blank\">here</a>. </p>\n<p>The main idea is mentioned in the abstract: </p>\n<blockquote>\n  <p>A sliding window operation was defined to crop each original image to smaller images. A convolutional neural network (CNN) was trained with cropped images as inputs and corresponding labels as output. Using this model, an image processing routine was developed to scan the test images to segment the GS glomeruli.</p>\n</blockquote>\n<p>Good luck!</p>",
  "messages": [
    {
      "id": "1081767",
      "postDate": "11/17/2020 09:20:32",
      "content": "<p>In case, someone missed it, the paper linked under the data section of this competition actually provides a good summary on how to start with this competition. </p>\n<p>You can reference it <a href=\"https://www.kireports.org/article/S2468-0249%2819%2930155-X/abstract\" target=\"_blank\">here</a>. </p>\n<p>The main idea is mentioned in the abstract: </p>\n<blockquote>\n  <p>A sliding window operation was defined to crop each original image to smaller images. A convolutional neural network (CNN) was trained with cropped images as inputs and corresponding labels as output. Using this model, an image processing routine was developed to scan the test images to segment the GS glomeruli.</p>\n</blockquote>\n<p>Good luck!</p>",
      "rawMarkdown": "In case, someone missed it, the paper linked under the data section of this competition actually provides a good summary on how to start with this competition. \n\nYou can reference it [here](https://www.kireports.org/article/S2468-0249%2819%2930155-X/abstract). \n\nThe main idea is mentioned in the abstract: \n> A sliding window operation was defined to crop each original image to smaller images. A convolutional neural network (CNN) was trained with cropped images as inputs and corresponding labels as output. Using this model, an image processing routine was developed to scan the test images to segment the GS glomeruli.\n\nGood luck!",
      "votes": null
    },
    {
      "id": "1082031",
      "postDate": "11/17/2020 14:25:42",
      "content": "<blockquote>\n  <p>A sliding window operation was defined to crop each original image to smaller images</p>\n</blockquote>\n<p>That sounds like the description of a convolution, a sliding window. I'm going to read the paper, thanks for resume the main idea.</p>",
      "rawMarkdown": "> A sliding window operation was defined to crop each original image to smaller images\n\nThat sounds like the description of a convolution, a sliding window. I'm going to read the paper, thanks for resume the main idea.",
      "votes": null
    },
    {
      "id": "1083268",
      "postDate": "11/18/2020 19:24:33",
      "content": "<p>Also see: <a href=\"https://www.kaggle.com/leahscherschel/glomeruli-detection-related-works\" target=\"_blank\">https://www.kaggle.com/leahscherschel/glomeruli-detection-related-works</a></p>",
      "rawMarkdown": "Also see: https://www.kaggle.com/leahscherschel/glomeruli-detection-related-works",
      "votes": null
    }
  ],
  "comments": [
    {
      "id": 1082031,
      "author_name": "hiramcho",
      "author_url": "",
      "post_date": "11/17/2020 14:25:42",
      "content": "<blockquote>\n  <p>A sliding window operation was defined to crop each original image to smaller images</p>\n</blockquote>\n<p>That sounds like the description of a convolution, a sliding window. I'm going to read the paper, thanks for resume the main idea.</p>",
      "votes": null,
      "replies": []
    },
    {
      "id": 1083268,
      "author_name": "leahscherschel",
      "author_url": "",
      "post_date": "11/18/2020 19:24:33",
      "content": "<p>Also see: <a href=\"https://www.kaggle.com/leahscherschel/glomeruli-detection-related-works\" target=\"_blank\">https://www.kaggle.com/leahscherschel/glomeruli-detection-related-works</a></p>",
      "votes": null,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "1081767": "In case, someone missed it, the paper linked under the data section of this competition actually provides a good summary on how to start with this competition. \n\nYou can reference it [here](https://www.kireports.org/article/S2468-0249%2819%2930155-X/abstract). \n\nThe main idea is mentioned in the abstract: \n> A sliding window operation was defined to crop each original image to smaller images. A convolutional neural network (CNN) was trained with cropped images as inputs and corresponding labels as output. Using this model, an image processing routine was developed to scan the test images to segment the GS glomeruli.\n\nGood luck!",
    "1082031": "> A sliding window operation was defined to crop each original image to smaller images\n\nThat sounds like the description of a convolution, a sliding window. I'm going to read the paper, thanks for resume the main idea.",
    "1083268": "Also see: https://www.kaggle.com/leahscherschel/glomeruli-detection-related-works"
  },
  "source": "meta"
}