After a year of hard work across Slack, Zoom, Google Meet, and many other tools of the post-pandemic research world, we are happy to announce that our paper on the PANDA challenge has been published in Nature Medicine. In the paper, we took a deep dive into all the solutions, tested the methods to see if they generalize well to unseen data, and performed a comparison with pathologists.
\nFrom the competition, the following teams have participated and presented their results in the paper:
\nDmitry A. Grechka, rähmä.ai, Aksell, KovaLOVE v2, UCLA Computational Diagnostics Lab, Iafoss, vanda, Manuel Campos, NS Pathology, ChienYiChi, ctrasd123, BarelyBears, Kiminya, Save The Prostate, PND
\nPublication
\nThe paper is now available online as an Open Access publication at Nature Medicine. You can read the full paper and all results here:
https://www.nature.com/articles/s41591-021-01620-2
\n\n\nBulten, W., Kartasalo, K., Chen, PH.C. et al. Artificial intelligence for diagnosis and Gleason grading of prostate cancer: the PANDA challenge. Nat Med (2022). https://doi.org/10.1038/s41591-021-01620-2
\n
A quick summary of the paper with some of the main results are also shown in this blog post: https://www.wouterbulten.nl/blog/tech/panda-challenge/
\nThank you
\nThe paper concludes a project that took over two years with many researchers, pathologists, AI developers, and others contributing.
\nWe, as organizers, would again like to congratulate all the teams who participated in the challenge. We are delighted to see that so many teams joined and achieved high scores on the leaderboard. Our special thanks go to all the teams that contributed extra time to the scientific part!
Dataset embargo lifted
\nWith the paper's publication, the embargo on the data is now lifted. If you want, you can now use the dataset for further scientific work and publish your results on the dataset. If you do so, please take the license (CC BY-SA-NC 4.0) into account and make sure you cite the PANDA paper. We are looking forward to seeing new scientific projects coming out of this dataset!
For any questions, don't hesitate to reach out! On behalf of the full team,
\nWouter
", "messages": [ { "id": 1649740, "postDate": "2022-01-14T14:53:05.170Z", "content": "After a year of hard work across Slack, Zoom, Google Meet, and many other tools of the post-pandemic research world, we are happy to announce that our paper on the PANDA challenge has been published in Nature Medicine. In the paper, we took a deep dive into all the solutions, tested the methods to see if they generalize well to unseen data, and performed a comparison with pathologists.
\nFrom the competition, the following teams have participated and presented their results in the paper:
\nDmitry A. Grechka, rähmä.ai, Aksell, KovaLOVE v2, UCLA Computational Diagnostics Lab, Iafoss, vanda, Manuel Campos, NS Pathology, ChienYiChi, ctrasd123, BarelyBears, Kiminya, Save The Prostate, PND
\nPublication
\nThe paper is now available online as an Open Access publication at Nature Medicine. You can read the full paper and all results here:
https://www.nature.com/articles/s41591-021-01620-2
\n\n\nBulten, W., Kartasalo, K., Chen, PH.C. et al. Artificial intelligence for diagnosis and Gleason grading of prostate cancer: the PANDA challenge. Nat Med (2022). https://doi.org/10.1038/s41591-021-01620-2
\n
A quick summary of the paper with some of the main results are also shown in this blog post: https://www.wouterbulten.nl/blog/tech/panda-challenge/
\nThank you
\nThe paper concludes a project that took over two years with many researchers, pathologists, AI developers, and others contributing.
\nWe, as organizers, would again like to congratulate all the teams who participated in the challenge. We are delighted to see that so many teams joined and achieved high scores on the leaderboard. Our special thanks go to all the teams that contributed extra time to the scientific part!
Dataset embargo lifted
\nWith the paper's publication, the embargo on the data is now lifted. If you want, you can now use the dataset for further scientific work and publish your results on the dataset. If you do so, please take the license (CC BY-SA-NC 4.0) into account and make sure you cite the PANDA paper. We are looking forward to seeing new scientific projects coming out of this dataset!
For any questions, don't hesitate to reach out! On behalf of the full team,
\nWouter
", "rawMarkdown": "After a year of hard work across Slack, Zoom, Google Meet, and many other tools of the post-pandemic research world, we are happy to announce that our paper on the PANDA challenge has been published in **Nature Medicine**. In the paper, we took a deep dive into all the solutions, tested the methods to see if they generalize well to unseen data, and performed a comparison with pathologists. \n\nFrom the competition, the following teams have participated and presented their results in the paper:\n\n*Dmitry A. Grechka, rähmä.ai, Aksell, KovaLOVE v2, UCLA Computational Diagnostics Lab, Iafoss, vanda, Manuel Campos, NS Pathology, ChienYiChi, ctrasd123, BarelyBears, Kiminya, Save The Prostate, PND*\n\n**Publication**\nThe paper is now available online as an Open Access publication at Nature Medicine. You can read the full paper and all results here:\n\nhttps://www.nature.com/articles/s41591-021-01620-2\n\n> Bulten, W., Kartasalo, K., Chen, PH.C. et al. Artificial intelligence for diagnosis and Gleason grading of prostate cancer: the PANDA challenge. Nat Med (2022). https://doi.org/10.1038/s41591-021-01620-2\n\nA quick summary of the paper with some of the main results are also shown in this blog post: https://www.wouterbulten.nl/blog/tech/panda-challenge/\n\n**Thank you**\nThe paper concludes a project that took over two years with many researchers, pathologists, AI developers, and others contributing. \nWe, as organizers, would again like to congratulate all the teams who participated in the challenge. We are delighted to see that so many teams joined and achieved high scores on the leaderboard. Our special thanks go to all the teams that contributed extra time to the scientific part! \n\n**Dataset embargo lifted**\nWith the paper's publication, the embargo on the data is now lifted. If you want, you can now use the dataset for further scientific work and publish your results on the dataset. If you do so, please take the license (CC BY-SA-NC 4.0) into account and make sure you cite the PANDA paper. We are looking forward to seeing new scientific projects coming out of this dataset!\n\nFor any questions, don't hesitate to reach out! On behalf of the full team,\n\nWouter\n\n", "votes": 20 }, { "id": 1847234, "postDate": "2022-07-07T18:23:14.727Z", "content": "congratulations, i came upon this topic from that nature article!
", "rawMarkdown": "congratulations, i came upon this topic from that nature article!", "votes": 1 }, { "id": 1707622, "postDate": "2022-02-28T16:08:02.523Z", "content": "Amazing!! I will be presenting it this week in our journal club!!
", "rawMarkdown": "Amazing!! I will be presenting it this week in our journal club!! ", "votes": 1 }, { "id": 1805377, "postDate": "2022-05-30T05:18:55.423Z", "content": "Thank you for your outstanding devotion!
", "rawMarkdown": "Thank you for your outstanding devotion!" }, { "id": 1792305, "postDate": "2022-05-16T20:04:07.417Z", "content": "Hi Wouter,
\nI've enjoyed following your research and reading the PANDA paper in Nature. Are any of the external and internal validation, or performance evaluation datasets available for download for our own performance evaluation.
\nThank you,
\nRyan
\nUSF Comp. Science & Engineering
\nTampa, FL
Hi, the current plan is to keep the evaluation datasets private. This allows people to still submit late submissions and get scores on the public and private sets - keeping PANDA useful for algorithm benchmarking.
", "rawMarkdown": "Hi, the current plan is to keep the evaluation datasets private. This allows people to still submit late submissions and get scores on the public and private sets - keeping PANDA useful for algorithm benchmarking. " } ] } ], "comments": [ { "id": 1847234, "author_name": "Valerio Zhang", "author_url": "", "post_date": "2022-07-07T18:23:14.727000", "content": "congratulations, i came upon this topic from that nature article!
", "votes": 1, "replies": [] }, { "id": 1707622, "author_name": "AhmadAlShammari", "author_url": "", "post_date": "2022-02-28T16:08:02.523000", "content": "Amazing!! I will be presenting it this week in our journal club!!
", "votes": 1, "replies": [] }, { "id": 1805377, "author_name": "Noel Tong", "author_url": "", "post_date": "2022-05-30T05:18:55.423000", "content": "Thank you for your outstanding devotion!
", "votes": 0, "replies": [] }, { "id": 1792305, "author_name": "RyanFog", "author_url": "", "post_date": "2022-05-16T20:04:07.417000", "content": "Hi Wouter,
\nI've enjoyed following your research and reading the PANDA paper in Nature. Are any of the external and internal validation, or performance evaluation datasets available for download for our own performance evaluation.
\nThank you,
\nRyan
\nUSF Comp. Science & Engineering
\nTampa, FL
Hi, the current plan is to keep the evaluation datasets private. This allows people to still submit late submissions and get scores on the public and private sets - keeping PANDA useful for algorithm benchmarking.
", "votes": 0, "replies": [] } ] } ], "raw_markdown_by_id": { "1649740": "After a year of hard work across Slack, Zoom, Google Meet, and many other tools of the post-pandemic research world, we are happy to announce that our paper on the PANDA challenge has been published in **Nature Medicine**. In the paper, we took a deep dive into all the solutions, tested the methods to see if they generalize well to unseen data, and performed a comparison with pathologists. \n\nFrom the competition, the following teams have participated and presented their results in the paper:\n\n*Dmitry A. Grechka, rähmä.ai, Aksell, KovaLOVE v2, UCLA Computational Diagnostics Lab, Iafoss, vanda, Manuel Campos, NS Pathology, ChienYiChi, ctrasd123, BarelyBears, Kiminya, Save The Prostate, PND*\n\n**Publication**\nThe paper is now available online as an Open Access publication at Nature Medicine. You can read the full paper and all results here:\n\nhttps://www.nature.com/articles/s41591-021-01620-2\n\n> Bulten, W., Kartasalo, K., Chen, PH.C. et al. Artificial intelligence for diagnosis and Gleason grading of prostate cancer: the PANDA challenge. Nat Med (2022). https://doi.org/10.1038/s41591-021-01620-2\n\nA quick summary of the paper with some of the main results are also shown in this blog post: https://www.wouterbulten.nl/blog/tech/panda-challenge/\n\n**Thank you**\nThe paper concludes a project that took over two years with many researchers, pathologists, AI developers, and others contributing. \nWe, as organizers, would again like to congratulate all the teams who participated in the challenge. We are delighted to see that so many teams joined and achieved high scores on the leaderboard. Our special thanks go to all the teams that contributed extra time to the scientific part! \n\n**Dataset embargo lifted**\nWith the paper's publication, the embargo on the data is now lifted. If you want, you can now use the dataset for further scientific work and publish your results on the dataset. If you do so, please take the license (CC BY-SA-NC 4.0) into account and make sure you cite the PANDA paper. We are looking forward to seeing new scientific projects coming out of this dataset!\n\nFor any questions, don't hesitate to reach out! On behalf of the full team,\n\nWouter\n\n", "1847234": "congratulations, i came upon this topic from that nature article!", "1707622": "Amazing!! I will be presenting it this week in our journal club!! ", "1805377": "Thank you for your outstanding devotion!", "1792305": "Hi Wouter, \n\nI've enjoyed following your research and reading the PANDA paper in Nature. Are any of the external and internal validation, or performance evaluation datasets available for download for our own performance evaluation.\n\nThank you,\nRyan\nUSF Comp. Science & Engineering\nTampa, FL\n" } }