{
  "id": 146364,
  "title": "PanNuke dataset available for everyone",
  "url": "/competitions/prostate-cancer-grade-assessment/discussion/146364",
  "author_name": "Larxel",
  "post_date": "2020-04-26T22:00:02.356000",
  "votes": 35,
  "comment_count": 7,
  "views": 0,
  "content": "<p>Hello everyone</p>\n\n<p>I posted the <strong>PanNuke dataset</strong> here on kaggle, with over <strong>200k nuclei annotated for both classification and instance segmentation</strong>.\nIt has 19 tissue types, including PCa.\nAll credits are due to the authors of the dataset (see references)</p>\n\n<p>Kaggle has a limit of 20gb per dataset, so I had to split it into 3 parts:\n- <a href=\"https://www.kaggle.com/andrewmvd/cancer-inst-segmentation-and-classification\">Part 1</a>\n- <a href=\"https://www.kaggle.com/andrewmvd/cancer-instance-segmentation-and-classification-2\">Part 2</a>\n- <a href=\"https://www.kaggle.com/andrewmvd/cancer-instance-segmentation-and-classification-3\">Part 3</a></p>\n\n<p>Hope you guys kick some cancer ass\nCheers</p>\n\n<p>References:\n@article{gamper2020pannuke,\n  title={PanNuke Dataset Extension, Insights and Baselines},\n  author={Gamper, Jevgenij and Koohbanani, Navid Alemi and Graham, Simon and Jahanifar, Mostafa and Benet, Ksenija and Khurram, Syed Ali and Azam, Ayesha and Hewitt, Katherine and Rajpoot, Nasir},\n  journal={arXiv preprint <a href=\"https://arxiv.org/abs/2003.10778\">arXiv:2003.10778</a>},\n  year={2020}\n}</p>\n\n<p>@inproceedings{gamper2019pannuke,\n  title={Pannuke: An open pan-cancer histology dataset for nuclei instance segmentation and classification},\n  author={Gamper, Jevgenij and Koohbanani, Navid Alemi and Benet, Ksenija and Khuram, Ali and Rajpoot, Nasir},\n  booktitle={European Congress on Digital Pathology},\n  pages={11--19},\n  year={2019},\n  organization={Springer}\n}</p>\n\n<p>Dataset License: CC BY NC SA 4.0</p>",
  "messages": [
    {
      "id": 822405,
      "postDate": "2020-04-26T22:00:02.357Z",
      "content": "<p>Hello everyone</p>\n\n<p>I posted the <strong>PanNuke dataset</strong> here on kaggle, with over <strong>200k nuclei annotated for both classification and instance segmentation</strong>.\nIt has 19 tissue types, including PCa.\nAll credits are due to the authors of the dataset (see references)</p>\n\n<p>Kaggle has a limit of 20gb per dataset, so I had to split it into 3 parts:\n- <a href=\"https://www.kaggle.com/andrewmvd/cancer-inst-segmentation-and-classification\">Part 1</a>\n- <a href=\"https://www.kaggle.com/andrewmvd/cancer-instance-segmentation-and-classification-2\">Part 2</a>\n- <a href=\"https://www.kaggle.com/andrewmvd/cancer-instance-segmentation-and-classification-3\">Part 3</a></p>\n\n<p>Hope you guys kick some cancer ass\nCheers</p>\n\n<p>References:\n@article{gamper2020pannuke,\n  title={PanNuke Dataset Extension, Insights and Baselines},\n  author={Gamper, Jevgenij and Koohbanani, Navid Alemi and Graham, Simon and Jahanifar, Mostafa and Benet, Ksenija and Khurram, Syed Ali and Azam, Ayesha and Hewitt, Katherine and Rajpoot, Nasir},\n  journal={arXiv preprint <a href=\"https://arxiv.org/abs/2003.10778\">arXiv:2003.10778</a>},\n  year={2020}\n}</p>\n\n<p>@inproceedings{gamper2019pannuke,\n  title={Pannuke: An open pan-cancer histology dataset for nuclei instance segmentation and classification},\n  author={Gamper, Jevgenij and Koohbanani, Navid Alemi and Benet, Ksenija and Khuram, Ali and Rajpoot, Nasir},\n  booktitle={European Congress on Digital Pathology},\n  pages={11--19},\n  year={2019},\n  organization={Springer}\n}</p>\n\n<p>Dataset License: CC BY NC SA 4.0</p>",
      "rawMarkdown": "Hello everyone\n\nI posted the **PanNuke dataset** here on kaggle, with over **200k nuclei annotated for both classification and instance segmentation**.\nIt has 19 tissue types, including PCa.\nAll credits are due to the authors of the dataset (see references)\n\nKaggle has a limit of 20gb per dataset, so I had to split it into 3 parts:\n- [Part 1](https://www.kaggle.com/andrewmvd/cancer-inst-segmentation-and-classification)\n- [Part 2](https://www.kaggle.com/andrewmvd/cancer-instance-segmentation-and-classification-2)\n- [Part 3](https://www.kaggle.com/andrewmvd/cancer-instance-segmentation-and-classification-3)\n\n\nHope you guys kick some cancer ass\nCheers\n\n\nReferences:\n@article{gamper2020pannuke,\n  title={PanNuke Dataset Extension, Insights and Baselines},\n  author={Gamper, Jevgenij and Koohbanani, Navid Alemi and Graham, Simon and Jahanifar, Mostafa and Benet, Ksenija and Khurram, Syed Ali and Azam, Ayesha and Hewitt, Katherine and Rajpoot, Nasir},\n  journal={arXiv preprint [arXiv:2003.10778](https://arxiv.org/abs/2003.10778)},\n  year={2020}\n}\n\n@inproceedings{gamper2019pannuke,\n  title={Pannuke: An open pan-cancer histology dataset for nuclei instance segmentation and classification},\n  author={Gamper, Jevgenij and Koohbanani, Navid Alemi and Benet, Ksenija and Khuram, Ali and Rajpoot, Nasir},\n  booktitle={European Congress on Digital Pathology},\n  pages={11--19},\n  year={2019},\n  organization={Springer}\n}\n\nDataset License: CC BY NC SA 4.0",
      "votes": 35
    },
    {
      "id": 855799,
      "postDate": "2020-05-21T08:15:26.977Z",
      "content": "<p>If you want to use a pre-trained model that will work out of the box on this dataset, you can refer to:</p>\n\n<p><a href=\"https://github.com/simongraham/hovernet_inference\">https://github.com/simongraham/hovernet_inference</a></p>\n\n<p>:)</p>",
      "rawMarkdown": "If you want to use a pre-trained model that will work out of the box on this dataset, you can refer to:\n\nhttps://github.com/simongraham/hovernet_inference\n\n:)",
      "votes": 10,
      "replies": [
        {
          "id": 855821,
          "postDate": "2020-05-21T08:50:40.743Z",
          "content": "<p>The model is simple to use with easy explained steps</p>",
          "rawMarkdown": "The model is simple to use with easy explained steps",
          "votes": 5
        }
      ]
    },
    {
      "id": 822824,
      "postDate": "2020-04-27T07:03:23.590Z",
      "content": "<p>Really appreciate your work.. Thanks.! </p>",
      "rawMarkdown": "Really appreciate your work.. Thanks.! ",
      "votes": 5,
      "replies": [
        {
          "id": 823236,
          "postDate": "2020-04-27T14:04:00.047Z",
          "content": "<p>You are welcome!</p>",
          "rawMarkdown": "You are welcome!",
          "votes": 1
        }
      ]
    },
    {
      "id": 822454,
      "postDate": "2020-04-26T23:02:20.203Z",
      "content": "<p>Hey <a href=\"/andrewmvd\">@andrewmvd</a> thanks a lot, I was looking for more images.</p>",
      "rawMarkdown": "Hey @andrewmvd thanks a lot, I was looking for more images.",
      "votes": 3,
      "replies": [
        {
          "id": 822456,
          "postDate": "2020-04-26T23:12:35.973Z",
          "content": "<p>You are welcome, if I find more I'll make it available</p>\n\n<p>Cheers!</p>",
          "rawMarkdown": "You are welcome, if I find more I'll make it available\n\nCheers!",
          "votes": 3
        }
      ]
    },
    {
      "id": 855881,
      "postDate": "2020-05-21T09:55:59.830Z",
      "content": "<p>this saved me, thanks!</p>",
      "rawMarkdown": "this saved me, thanks!\n",
      "votes": 3
    }
  ],
  "comments": [
    {
      "id": 855799,
      "author_name": "SimonGraham",
      "author_url": "",
      "post_date": "2020-05-21T08:15:26.977000",
      "content": "<p>If you want to use a pre-trained model that will work out of the box on this dataset, you can refer to:</p>\n\n<p><a href=\"https://github.com/simongraham/hovernet_inference\">https://github.com/simongraham/hovernet_inference</a></p>\n\n<p>:)</p>",
      "votes": 10,
      "replies": [
        {
          "id": 855821,
          "author_name": "Noorul Wahab",
          "author_url": "",
          "post_date": "2020-05-21T08:50:40.743000",
          "content": "<p>The model is simple to use with easy explained steps</p>",
          "votes": 5,
          "replies": []
        }
      ]
    },
    {
      "id": 822824,
      "author_name": "torch",
      "author_url": "",
      "post_date": "2020-04-27T07:03:23.590000",
      "content": "<p>Really appreciate your work.. Thanks.! </p>",
      "votes": 5,
      "replies": [
        {
          "id": 823236,
          "author_name": "Larxel",
          "author_url": "",
          "post_date": "2020-04-27T14:04:00.047000",
          "content": "<p>You are welcome!</p>",
          "votes": 1,
          "replies": []
        }
      ]
    },
    {
      "id": 822454,
      "author_name": "TheStoneMX",
      "author_url": "",
      "post_date": "2020-04-26T23:02:20.203000",
      "content": "<p>Hey <a href=\"/andrewmvd\">@andrewmvd</a> thanks a lot, I was looking for more images.</p>",
      "votes": 3,
      "replies": [
        {
          "id": 822456,
          "author_name": "Larxel",
          "author_url": "",
          "post_date": "2020-04-26T23:12:35.973000",
          "content": "<p>You are welcome, if I find more I'll make it available</p>\n\n<p>Cheers!</p>",
          "votes": 3,
          "replies": []
        }
      ]
    },
    {
      "id": 855881,
      "author_name": "LeonidZ.",
      "author_url": "",
      "post_date": "2020-05-21T09:55:59.830000",
      "content": "<p>this saved me, thanks!</p>",
      "votes": 3,
      "replies": []
    }
  ],
  "raw_markdown_by_id": {
    "822405": "Hello everyone\n\nI posted the **PanNuke dataset** here on kaggle, with over **200k nuclei annotated for both classification and instance segmentation**.\nIt has 19 tissue types, including PCa.\nAll credits are due to the authors of the dataset (see references)\n\nKaggle has a limit of 20gb per dataset, so I had to split it into 3 parts:\n- [Part 1](https://www.kaggle.com/andrewmvd/cancer-inst-segmentation-and-classification)\n- [Part 2](https://www.kaggle.com/andrewmvd/cancer-instance-segmentation-and-classification-2)\n- [Part 3](https://www.kaggle.com/andrewmvd/cancer-instance-segmentation-and-classification-3)\n\n\nHope you guys kick some cancer ass\nCheers\n\n\nReferences:\n@article{gamper2020pannuke,\n  title={PanNuke Dataset Extension, Insights and Baselines},\n  author={Gamper, Jevgenij and Koohbanani, Navid Alemi and Graham, Simon and Jahanifar, Mostafa and Benet, Ksenija and Khurram, Syed Ali and Azam, Ayesha and Hewitt, Katherine and Rajpoot, Nasir},\n  journal={arXiv preprint [arXiv:2003.10778](https://arxiv.org/abs/2003.10778)},\n  year={2020}\n}\n\n@inproceedings{gamper2019pannuke,\n  title={Pannuke: An open pan-cancer histology dataset for nuclei instance segmentation and classification},\n  author={Gamper, Jevgenij and Koohbanani, Navid Alemi and Benet, Ksenija and Khuram, Ali and Rajpoot, Nasir},\n  booktitle={European Congress on Digital Pathology},\n  pages={11--19},\n  year={2019},\n  organization={Springer}\n}\n\nDataset License: CC BY NC SA 4.0",
    "855799": "If you want to use a pre-trained model that will work out of the box on this dataset, you can refer to:\n\nhttps://github.com/simongraham/hovernet_inference\n\n:)",
    "822824": "Really appreciate your work.. Thanks.! ",
    "822454": "Hey @andrewmvd thanks a lot, I was looking for more images.",
    "855881": "this saved me, thanks!\n"
  }
}