{
  "id": 412233,
  "title": "Welcome to the HuBMAP - Hacking the Human Vasculature Competition!",
  "url": "/competitions/hubmap-hacking-the-human-vasculature/discussion/412233",
  "author_name": "Yashvardhan Jain",
  "post_date": "2023-05-22T19:56:53.550000",
  "votes": 31,
  "comment_count": 11,
  "views": 0,
  "content": "<p>Welcome to the HuBMAP - Hacking the Human Vasculature Competition!</p>\n<p>We are very much looking forward to the innovative solutions teams will generate to segment microvascular structures in histology images of human kidney tissue.</p>\n<p>To help you get started, please check out the following resources:</p>\n<ul>\n<li><a href=\"https://www.kaggle.com/competitions/hubmap-hacking-the-human-vasculature/overview/biological-overview\" target=\"_blank\">Biological Background Information</a></li>\n<li><a href=\"https://www.frontiersin.org/articles/10.3389/fcvm.2020.00029/full\" target=\"_blank\">Vasculature Common Coordinate Framework</a></li>\n<li><a href=\"https://www.kaggle.com/competitions/hubmap-kidney-segmentation\" target=\"_blank\">HuBMAP : Hacking the Kidney</a></li>\n<li><a href=\"https://www.kaggle.com/competitions/hubmap-organ-segmentation\" target=\"_blank\">HuBMAP + HPA : Hacking the Human Body</a></li>\n<li>Results papers (preprints) for previous HuBMAP competitions are <a href=\"https://www.biorxiv.org/content/10.1101/2023.01.05.522764v1\" target=\"_blank\">here</a> and <a href=\"https://www.biorxiv.org/content/10.1101/2021.11.09.467810v1.abstract\" target=\"_blank\">here</a></li>\n</ul>\n<p>Our team will be monitoring the discussion board and we look forward to hearing from you! Good luck and have fun!</p>\n<p>Yash<br>\nResearch Software Engineer - Machine Learning<br>\nIndiana University</p>",
  "messages": [
    {
      "id": 2269919,
      "postDate": "2023-05-22T19:56:53.550Z",
      "content": "<p>Welcome to the HuBMAP - Hacking the Human Vasculature Competition!</p>\n<p>We are very much looking forward to the innovative solutions teams will generate to segment microvascular structures in histology images of human kidney tissue.</p>\n<p>To help you get started, please check out the following resources:</p>\n<ul>\n<li><a href=\"https://www.kaggle.com/competitions/hubmap-hacking-the-human-vasculature/overview/biological-overview\" target=\"_blank\">Biological Background Information</a></li>\n<li><a href=\"https://www.frontiersin.org/articles/10.3389/fcvm.2020.00029/full\" target=\"_blank\">Vasculature Common Coordinate Framework</a></li>\n<li><a href=\"https://www.kaggle.com/competitions/hubmap-kidney-segmentation\" target=\"_blank\">HuBMAP : Hacking the Kidney</a></li>\n<li><a href=\"https://www.kaggle.com/competitions/hubmap-organ-segmentation\" target=\"_blank\">HuBMAP + HPA : Hacking the Human Body</a></li>\n<li>Results papers (preprints) for previous HuBMAP competitions are <a href=\"https://www.biorxiv.org/content/10.1101/2023.01.05.522764v1\" target=\"_blank\">here</a> and <a href=\"https://www.biorxiv.org/content/10.1101/2021.11.09.467810v1.abstract\" target=\"_blank\">here</a></li>\n</ul>\n<p>Our team will be monitoring the discussion board and we look forward to hearing from you! Good luck and have fun!</p>\n<p>Yash<br>\nResearch Software Engineer - Machine Learning<br>\nIndiana University</p>",
      "rawMarkdown": "Welcome to the HuBMAP - Hacking the Human Vasculature Competition!\n\nWe are very much looking forward to the innovative solutions teams will generate to segment microvascular structures in histology images of human kidney tissue.\n\nTo help you get started, please check out the following resources:\n- [Biological Background Information](https://www.kaggle.com/competitions/hubmap-hacking-the-human-vasculature/overview/biological-overview)\n- [Vasculature Common Coordinate Framework](https://www.frontiersin.org/articles/10.3389/fcvm.2020.00029/full)\n- [HuBMAP : Hacking the Kidney](https://www.kaggle.com/competitions/hubmap-kidney-segmentation)\n- [HuBMAP + HPA : Hacking the Human Body](https://www.kaggle.com/competitions/hubmap-organ-segmentation)\n- Results papers (preprints) for previous HuBMAP competitions are [here](https://www.biorxiv.org/content/10.1101/2023.01.05.522764v1) and [here](https://www.biorxiv.org/content/10.1101/2021.11.09.467810v1.abstract)\n\n\nOur team will be monitoring the discussion board and we look forward to hearing from you! Good luck and have fun!\n\nYash\nResearch Software Engineer - Machine Learning\nIndiana University",
      "votes": 31
    },
    {
      "id": 2269921,
      "postDate": "2023-05-22T20:02:05.267Z",
      "content": "<p>Welcome to the competition, everyone! Best of luck!</p>\n<p>-Kate </p>\n<p>Research Analyst &amp; Anatomist</p>\n<p>Indiana University </p>",
      "rawMarkdown": "Welcome to the competition, everyone! Best of luck!\n\n-Kate \n\nResearch Analyst & Anatomist\n\nIndiana University ",
      "votes": 6,
      "replies": [
        {
          "id": 2366299,
          "postDate": "2023-07-30T22:04:06.337Z",
          "rawMarkdown": "",
          "isDeleted": true
        }
      ]
    },
    {
      "id": 2304233,
      "postDate": "2023-06-15T20:21:32.540Z",
      "content": "<p>Thanks for this new HuBMAP competition and the many resources. Hopefully something useful comes out of it. </p>",
      "rawMarkdown": "Thanks for this new HuBMAP competition and the many resources. Hopefully something useful comes out of it. ",
      "votes": 1
    },
    {
      "id": 2295817,
      "postDate": "2023-06-11T09:16:27.760Z",
      "content": "<h1>Hello, I have been trying to find an answer to my question for a long time, but I could not. I apologize if the question seems stupid 👀</h1>\n<p><strong>I have good experience in segmentation however this is my first competition.</strong></p>\n<h2>Test data</h2>\n<p><strong>It is also not entirely clear to me how we send the result, but there is a mask. In the forecast line we have some kind of line. I'm assuming it's some sort of coded information.</strong></p>\n<h2>Internet off in notebook</h2>\n<p><strong>In the requirements for the competition, the notebook must be disconnected from the Internet. Do I understand correctly that, in fact, the competition requires that we use only pre-installed libraries and not download third-party repositories?</strong></p>\n<p>I will be immensely grateful for your help 💚💚💚</p>",
      "rawMarkdown": "# Hello, I have been trying to find an answer to my question for a long time, but I could not. I apologize if the question seems stupid 👀\n**I have good experience in segmentation however this is my first competition.**\n\n## Test data\n**It is also not entirely clear to me how we send the result, but there is a mask. In the forecast line we have some kind of line. I'm assuming it's some sort of coded information.**\n\n## Internet off in notebook\n**In the requirements for the competition, the notebook must be disconnected from the Internet. Do I understand correctly that, in fact, the competition requires that we use only pre-installed libraries and not download third-party repositories?**\n\nI will be immensely grateful for your help 💚💚💚",
      "votes": 2,
      "replies": [
        {
          "id": 2304238,
          "postDate": "2023-06-15T20:27:38.297Z",
          "content": "<p>Here are some answers, please ask if you need more details:</p>\n<ol>\n<li>The <a href=\"https://en.wikipedia.org/wiki/Run-length_encoding\" target=\"_blank\"><strong>RLE</strong></a> mask (which is already an encoding of the pixels) is <strong>encoded</strong> using base64 and compressed with zlib. This transforms start and end RLE mask pixels into a string. For example: <strong>eNoLTDAwyrM3yI/PMwcAE94DZA==</strong></li>\n<li>Indeed, internet is disabled during the inference (and only the inference, you can use it for training). That doesn't mean you can't use libraries, you will need to install these \"offline\". For that, you can use various available datasets that contain the libraries you need or create your own dataset. If this point isn't clear, please ask for more details.</li>\n</ol>\n<p>I hope this helps a bit!</p>",
          "rawMarkdown": "Here are some answers, please ask if you need more details:\n\n1. The [**RLE**](https://en.wikipedia.org/wiki/Run-length_encoding) mask (which is already an encoding of the pixels) is **encoded** using base64 and compressed with zlib. This transforms start and end RLE mask pixels into a string. For example: **eNoLTDAwyrM3yI/PMwcAE94DZA==**\n2.  Indeed, internet is disabled during the inference (and only the inference, you can use it for training). That doesn't mean you can't use libraries, you will need to install these \"offline\". For that, you can use various available datasets that contain the libraries you need or create your own dataset. If this point isn't clear, please ask for more details.\n\nI hope this helps a bit!",
          "votes": 2
        }
      ]
    },
    {
      "id": 2273675,
      "postDate": "2023-05-25T10:29:27.543Z",
      "content": "<p>Sir/Madam, I have only the MATLAB can I do the codding in this software</p>",
      "rawMarkdown": "Sir/Madam, I have only the MATLAB can I do the codding in this software",
      "replies": [
        {
          "id": 2304239,
          "postDate": "2023-06-15T20:30:06.917Z",
          "content": "<p>I think you can (at least for the training part) but it isn't recommended (MATLAB computer vision ecosystem is less developed than the Python one as far as I know). I would suggest using Python. It is free and you can pick up the basics pretty quickly since you know MATLAB.</p>",
          "rawMarkdown": "I think you can (at least for the training part) but it isn't recommended (MATLAB computer vision ecosystem is less developed than the Python one as far as I know). I would suggest using Python. It is free and you can pick up the basics pretty quickly since you know MATLAB."
        }
      ]
    },
    {
      "id": 2356775,
      "postDate": "2023-07-24T11:18:24.857Z",
      "content": "<p>So I have a question about the submission format. Say I have two instances of blood vessels, then is it supposed to be two different encoded strings for both? Or can we package all blood vessels into one numpy array and convert it to a single encoded string?</p>\n<p>Please respond, as I really cant afford to get the submission format wrong. Thank you!</p>",
      "rawMarkdown": "So I have a question about the submission format. Say I have two instances of blood vessels, then is it supposed to be two different encoded strings for both? Or can we package all blood vessels into one numpy array and convert it to a single encoded string?\n\nPlease respond, as I really cant afford to get the submission format wrong. Thank you!"
    },
    {
      "id": 2355655,
      "postDate": "2023-07-23T13:42:38.080Z",
      "content": "<p>I have a question regarding the coordinate system used in the json file. Are the points given simply the border of the polygon, or is the whole blood vessel supposed to look like a string?</p>",
      "rawMarkdown": "I have a question regarding the coordinate system used in the json file. Are the points given simply the border of the polygon, or is the whole blood vessel supposed to look like a string?"
    },
    {
      "id": 2366297,
      "postDate": "2023-07-30T22:01:31.417Z",
      "rawMarkdown": "",
      "isDeleted": true
    },
    {
      "id": 2357153,
      "postDate": "2023-07-24T15:13:43.580Z",
      "content": "<p>Thanks for HuBMAP competition, nice.</p>",
      "rawMarkdown": "Thanks for HuBMAP competition, nice."
    }
  ],
  "comments": [
    {
      "id": 2269921,
      "author_name": "Katherine Gustilo",
      "author_url": "",
      "post_date": "2023-05-22T20:02:05.267000",
      "content": "<p>Welcome to the competition, everyone! Best of luck!</p>\n<p>-Kate </p>\n<p>Research Analyst &amp; Anatomist</p>\n<p>Indiana University </p>",
      "votes": 6,
      "replies": [
        {
          "id": 2366299,
          "author_name": "",
          "author_url": "",
          "post_date": "2023-07-30T22:04:06.337000",
          "content": "",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 2304233,
      "author_name": "Yassine Alouini",
      "author_url": "",
      "post_date": "2023-06-15T20:21:32.540000",
      "content": "<p>Thanks for this new HuBMAP competition and the many resources. Hopefully something useful comes out of it. </p>",
      "votes": 1,
      "replies": []
    },
    {
      "id": 2295817,
      "author_name": "",
      "author_url": "",
      "post_date": "2023-06-11T09:16:27.760000",
      "content": "<h1>Hello, I have been trying to find an answer to my question for a long time, but I could not. I apologize if the question seems stupid 👀</h1>\n<p><strong>I have good experience in segmentation however this is my first competition.</strong></p>\n<h2>Test data</h2>\n<p><strong>It is also not entirely clear to me how we send the result, but there is a mask. In the forecast line we have some kind of line. I'm assuming it's some sort of coded information.</strong></p>\n<h2>Internet off in notebook</h2>\n<p><strong>In the requirements for the competition, the notebook must be disconnected from the Internet. Do I understand correctly that, in fact, the competition requires that we use only pre-installed libraries and not download third-party repositories?</strong></p>\n<p>I will be immensely grateful for your help 💚💚💚</p>",
      "votes": 2,
      "replies": [
        {
          "id": 2304238,
          "author_name": "Yassine Alouini",
          "author_url": "",
          "post_date": "2023-06-15T20:27:38.297000",
          "content": "<p>Here are some answers, please ask if you need more details:</p>\n<ol>\n<li>The <a href=\"https://en.wikipedia.org/wiki/Run-length_encoding\" target=\"_blank\"><strong>RLE</strong></a> mask (which is already an encoding of the pixels) is <strong>encoded</strong> using base64 and compressed with zlib. This transforms start and end RLE mask pixels into a string. For example: <strong>eNoLTDAwyrM3yI/PMwcAE94DZA==</strong></li>\n<li>Indeed, internet is disabled during the inference (and only the inference, you can use it for training). That doesn't mean you can't use libraries, you will need to install these \"offline\". For that, you can use various available datasets that contain the libraries you need or create your own dataset. If this point isn't clear, please ask for more details.</li>\n</ol>\n<p>I hope this helps a bit!</p>",
          "votes": 2,
          "replies": []
        }
      ]
    },
    {
      "id": 2273675,
      "author_name": "SRI MURALI V .K",
      "author_url": "",
      "post_date": "2023-05-25T10:29:27.543000",
      "content": "<p>Sir/Madam, I have only the MATLAB can I do the codding in this software</p>",
      "votes": 0,
      "replies": [
        {
          "id": 2304239,
          "author_name": "Yassine Alouini",
          "author_url": "",
          "post_date": "2023-06-15T20:30:06.917000",
          "content": "<p>I think you can (at least for the training part) but it isn't recommended (MATLAB computer vision ecosystem is less developed than the Python one as far as I know). I would suggest using Python. It is free and you can pick up the basics pretty quickly since you know MATLAB.</p>",
          "votes": 0,
          "replies": []
        }
      ]
    },
    {
      "id": 2356775,
      "author_name": "Malhar_k",
      "author_url": "",
      "post_date": "2023-07-24T11:18:24.857000",
      "content": "<p>So I have a question about the submission format. Say I have two instances of blood vessels, then is it supposed to be two different encoded strings for both? Or can we package all blood vessels into one numpy array and convert it to a single encoded string?</p>\n<p>Please respond, as I really cant afford to get the submission format wrong. Thank you!</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2355655,
      "author_name": "Malhar_k",
      "author_url": "",
      "post_date": "2023-07-23T13:42:38.080000",
      "content": "<p>I have a question regarding the coordinate system used in the json file. Are the points given simply the border of the polygon, or is the whole blood vessel supposed to look like a string?</p>",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2366297,
      "author_name": "",
      "author_url": "",
      "post_date": "2023-07-30T22:01:31.417000",
      "content": "",
      "votes": 0,
      "replies": []
    },
    {
      "id": 2357153,
      "author_name": "zeaulo",
      "author_url": "",
      "post_date": "2023-07-24T15:13:43.580000",
      "content": "<p>Thanks for HuBMAP competition, nice.</p>",
      "votes": 0,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "2269919": "Welcome to the HuBMAP - Hacking the Human Vasculature Competition!\n\nWe are very much looking forward to the innovative solutions teams will generate to segment microvascular structures in histology images of human kidney tissue.\n\nTo help you get started, please check out the following resources:\n- [Biological Background Information](https://www.kaggle.com/competitions/hubmap-hacking-the-human-vasculature/overview/biological-overview)\n- [Vasculature Common Coordinate Framework](https://www.frontiersin.org/articles/10.3389/fcvm.2020.00029/full)\n- [HuBMAP : Hacking the Kidney](https://www.kaggle.com/competitions/hubmap-kidney-segmentation)\n- [HuBMAP + HPA : Hacking the Human Body](https://www.kaggle.com/competitions/hubmap-organ-segmentation)\n- Results papers (preprints) for previous HuBMAP competitions are [here](https://www.biorxiv.org/content/10.1101/2023.01.05.522764v1) and [here](https://www.biorxiv.org/content/10.1101/2021.11.09.467810v1.abstract)\n\n\nOur team will be monitoring the discussion board and we look forward to hearing from you! Good luck and have fun!\n\nYash\nResearch Software Engineer - Machine Learning\nIndiana University",
    "2269921": "Welcome to the competition, everyone! Best of luck!\n\n-Kate \n\nResearch Analyst & Anatomist\n\nIndiana University ",
    "2304233": "Thanks for this new HuBMAP competition and the many resources. Hopefully something useful comes out of it. ",
    "2295817": "# Hello, I have been trying to find an answer to my question for a long time, but I could not. I apologize if the question seems stupid 👀\n**I have good experience in segmentation however this is my first competition.**\n\n## Test data\n**It is also not entirely clear to me how we send the result, but there is a mask. In the forecast line we have some kind of line. I'm assuming it's some sort of coded information.**\n\n## Internet off in notebook\n**In the requirements for the competition, the notebook must be disconnected from the Internet. Do I understand correctly that, in fact, the competition requires that we use only pre-installed libraries and not download third-party repositories?**\n\nI will be immensely grateful for your help 💚💚💚",
    "2273675": "Sir/Madam, I have only the MATLAB can I do the codding in this software",
    "2356775": "So I have a question about the submission format. Say I have two instances of blood vessels, then is it supposed to be two different encoded strings for both? Or can we package all blood vessels into one numpy array and convert it to a single encoded string?\n\nPlease respond, as I really cant afford to get the submission format wrong. Thank you!",
    "2355655": "I have a question regarding the coordinate system used in the json file. Are the points given simply the border of the polygon, or is the whole blood vessel supposed to look like a string?",
    "2366297": "",
    "2357153": "Thanks for HuBMAP competition, nice."
  }
}