![\"results-10-1.png\"](\"https://i.postimg.cc/LsfZGt7g/results-10-1.png\")
As per @yashvrdnjain's response, I wanted to compile all mentioned external sources here in one place. I will update the list below throughout the competition.
\nIf you have an external data source not listed below, please comment the details and I will add them to the table
\n[ORIGINAL POST]
\nHi all, I'm not sure how using external data works.
\nI haven't yet started to use external data but I'm sure others have.
\nUsually, when I use external data I share it with the community on an existing thread dedicated to this purpose. I'm not seeing anything like that so far in this competition.
Do we need to share those sources publically during the competition? Is there a place to do this?
\nEven if it's not mandated, I would/will encourage others to share the external data sources they are using. Either make a post detailing your sources or you can reply here with a comment?
\nFor reference here is the part of the rules I could find related to external data:
\n\n\nC. External Data. You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).
\n
All the best!
", "messages": [ { "id": "1836511", "postDate": "06/28/2022 18:04:16", "content": "As per @yashvrdnjain's response, I wanted to compile all mentioned external sources here in one place. I will update the list below throughout the competition.
\nIf you have an external data source not listed below, please comment the details and I will add them to the table
\n[ORIGINAL POST]
\nHi all, I'm not sure how using external data works.
\nI haven't yet started to use external data but I'm sure others have.
\nUsually, when I use external data I share it with the community on an existing thread dedicated to this purpose. I'm not seeing anything like that so far in this competition.
Do we need to share those sources publically during the competition? Is there a place to do this?
\nEven if it's not mandated, I would/will encourage others to share the external data sources they are using. Either make a post detailing your sources or you can reply here with a comment?
\nFor reference here is the part of the rules I could find related to external data:
\n\n\nC. External Data. You may use data other than the Competition Data (“External Data”) to develop and test your Submissions. However, you will ensure the External Data is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants. The ability to use External Data under this Section 7.C (External Data) does not limit your other obligations under these Competition Rules, including but not limited to Section 11 (Winners Obligations).
\n
All the best!
", "rawMarkdown": "## **[UPDATE]**\n\n***As per @yashvrdnjain's response, I wanted to compile all mentioned external sources here in one place. I will update the list below throughout the competition.***\n\n***If you have an external data source not listed below, please comment the details and I will add them to the table***\n\nAll participants must ensure the External Data they use is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants.
\nThank you for creating this thread. We encourage all participants to post their external data sources here or by creating a new discussion post.
I'll comb through the discussion posts that have already been posted and edit my main post to compile them. That should help get it started.
", "rawMarkdown": "I'll comb through the discussion posts that have already been posted and edit my main post to compile them. That should help get it started.", "votes": null }, { "id": "1842779", "postDate": "07/04/2022 09:26:23", "content": "Thanks for putting this together!
", "rawMarkdown": "Thanks for putting this together!", "votes": null }, { "id": "1842922", "postDate": "07/04/2022 11:53:12", "content": "There is also the cancer imaging archive, while it has a focus on cancerous tissue, it might also be used for unsupervised pretraining since it contains a lot of data 1.4 tb of prostate cancer
\n66.6gb of colon cancer and 21 gb of lung cancer
Overview: https://www.cancerimagingarchive.net/histopathology-imaging-on-tcia/
", "rawMarkdown": "There is also the cancer imaging archive, while it has a focus on cancerous tissue, it might also be used for unsupervised pretraining since it contains **a lot** of data [1.4 tb of prostate cancer](https://wiki.cancerimagingarchive.net/pages/viewpage.action?pageId=91357374) \n[66.6gb of colon cancer](https://wiki.cancerimagingarchive.net/display/Public/CPTAC-COAD) and [21 gb of lung cancer] (https://wiki.cancerimagingarchive.net/display/Public/Lung+Fused-CT-Pathology)\n\n[example image in viewer](https://pathdb.cancerimagingarchive.net/caMicroscope/apps/viewer/viewer.html?slideId=219753&mode=pathdb)\n\nOverview: https://www.cancerimagingarchive.net/histopathology-imaging-on-tcia/", "votes": null }, { "id": "1848486", "postDate": "07/08/2022 16:55:08", "content": "1.4 Tb isn't just a lot of data its tons of data the essenstial key point here is even if kagglers want to make some sense of this kind of data there isn't enough compute available it could take months to make sense of what they are actually trying to do
", "rawMarkdown": "1.4 Tb isn't just a lot of data its tons of data the essenstial key point here is even if kagglers want to make some sense of this kind of data there isn't enough compute available it could take months to make sense of what they are actually trying to do", "votes": null }, { "id": "1860453", "postDate": "07/18/2022 10:29:19", "content": "There are plenty of H&E stained images, i hope they are helpful to you, good luck
\nKidney: images and GT,
\n 1. (Hubmap): https://www.kaggle.com/competitions/hubmap-kidney-segmentation/data.
\n 2. CPTAC-CCRCC: https://wiki.cancerimagingarchive.net/display/Public/CPTAC-CCRCC
Large intestine: images
\n 1. TCGA COAD: https://wiki.cancerimagingarchive.net/display/Public/TCGA-COAD
\n 2. CPTAC-COAD: https://wiki.cancerimagingarchive.net/display/Public/CPTAC-COAD
Lung: images
\n 1. TCGA LUAD: https://wiki.cancerimagingarchive.net/display/Public/TCGA-LUAD
\n 2. TCGA LUSC: https://wiki.cancerimagingarchive.net/display/Public/TCGA-LUSC
\n 3. Lung-ACDC: https://acdc-lunghp.grand-challenge.org/Challenge/
\n 4. CPTAC-LSCC: https://wiki.cancerimagingarchive.net/display/Public/CPTAC-LSCC
\n 5. CPTAC-LUAD: https://wiki.cancerimagingarchive.net/display/Public/CPTAC-LUAD
Prostate
\n 1. TCGA-PRAD: https://wiki.cancerimagingarchive.net/display/Public/TCGA-PRAD
\n 2. PANDAs: https://www.kaggle.com/competitions/prostate-cancer-grade-assessment
\n 1. https://www.cancerimagingarchive.net/histopathology-imaging-on-tcia/
\n 2. https://wiki.cancerimagingarchive.net/display/Public/PROSTATE-MRI
Kidney + Large intestine + Lung + Spleen + Prostate:
\n (HPA) https://www.proteinatlas.org/about/download
while any tissue slide would help, but it is important to note:
\n\"All tissue data used in this competition is from healthy donors that pathologists identified as pathologically unremarkable tissue.\"
\nhence I would try non-cancer tissue first
", "rawMarkdown": "while any tissue slide would help, but it is important to note:\n\n\"All tissue data used in this competition is from healthy donors that pathologists identified as pathologically unremarkable tissue.\"\n\nhence I would try non-cancer tissue first", "votes": null }, { "id": "1866401", "postDate": "07/22/2022 12:54:58", "content": "Where are the annotations from the FTUs on the Hubmap Portal?
\nI cant find them or do they even exist?
Thanks and regards
", "rawMarkdown": "Where are the annotations from the FTUs on the Hubmap Portal? \nI cant find them or do they even exist?\n\nThanks and regards", "votes": null }, { "id": "1896346", "postDate": "08/12/2022 19:06:13", "content": "Regarding the mentioned HuBMAP - Hacking the Kidney competition, I created a simple kernel for loading the enormous images and exporting binary masks to a dataset - https://www.kaggle.com/datasets/jirkaborovec/hacking-the-kidney-annotation-masks-png-images
\n
\n", "rawMarkdown": "Regarding the mentioned HuBMAP - Hacking the Kidney competition, I created a simple kernel for loading the enormous images and exporting binary masks to a dataset - https://www.kaggle.com/datasets/jirkaborovec/hacking-the-kidney-annotation-masks-png-images\n\n[](https://postimg.cc/JyrnkkRR)\n\n> https://www.kaggle.com/code/jirkaborovec/hubmap-segm-eda-export-rle-mask", "votes": null }, { "id": "1922919", "postDate": "09/01/2022 20:23:31", "content": "https://www.kaggle.com/code/jirkaborovec/hubmap-segm-eda-export-rle-mask
\n
And the data here: https://cns-iu.github.io/ccf-research-kaggle-2021/
\nGtex: https://gtexportal.org/home/histologyPage
\nJust found dataset for prostate: https://www.wouterbulten.nl/blog/tech/peso-dataset-whole-slide-image-prosate-cancer/
", "rawMarkdown": "And the data here: https://cns-iu.github.io/ccf-research-kaggle-2021/\n\nGtex: https://gtexportal.org/home/histologyPage\n\nJust found dataset for prostate: https://www.wouterbulten.nl/blog/tech/peso-dataset-whole-slide-image-prosate-cancer/", "votes": null }, { "id": "1936994", "postDate": "09/13/2022 07:18:03", "content": "Very helpful! Thanks.
", "rawMarkdown": "Very helpful! Thanks.", "votes": null }, { "id": "1969253", "postDate": "10/03/2022 11:57:55", "content": "is there avaliable data set
", "rawMarkdown": "is there avaliable data set", "votes": null } ], "comments": [ { "id": 1839652, "author_name": "yashvrdnjain", "author_url": "", "post_date": "07/01/2022 15:38:11", "content": "All participants must ensure the External Data they use is publicly available and equally accessible to use by all participants of the Competition for purposes of the competition at no cost to the other participants.
\nThank you for creating this thread. We encourage all participants to post their external data sources here or by creating a new discussion post.
I'll comb through the discussion posts that have already been posted and edit my main post to compile them. That should help get it started.
", "votes": null, "replies": [] } ] }, { "id": 1842779, "author_name": "virajkadam", "author_url": "", "post_date": "07/04/2022 09:26:23", "content": "Thanks for putting this together!
", "votes": null, "replies": [] }, { "id": 1842922, "author_name": "theudas", "author_url": "", "post_date": "07/04/2022 11:53:12", "content": "There is also the cancer imaging archive, while it has a focus on cancerous tissue, it might also be used for unsupervised pretraining since it contains a lot of data 1.4 tb of prostate cancer
\n66.6gb of colon cancer and 21 gb of lung cancer
Overview: https://www.cancerimagingarchive.net/histopathology-imaging-on-tcia/
", "votes": null, "replies": [ { "id": 1848486, "author_name": "harshris21", "author_url": "", "post_date": "07/08/2022 16:55:08", "content": "1.4 Tb isn't just a lot of data its tons of data the essenstial key point here is even if kagglers want to make some sense of this kind of data there isn't enough compute available it could take months to make sense of what they are actually trying to do
", "votes": null, "replies": [] }, { "id": 1861364, "author_name": "hengck23", "author_url": "", "post_date": "07/19/2022 01:51:08", "content": "while any tissue slide would help, but it is important to note:
\n\"All tissue data used in this competition is from healthy donors that pathologists identified as pathologically unremarkable tissue.\"
\nhence I would try non-cancer tissue first
", "votes": null, "replies": [] } ] }, { "id": 1860453, "author_name": "lsl000ud", "author_url": "", "post_date": "07/18/2022 10:29:19", "content": "There are plenty of H&E stained images, i hope they are helpful to you, good luck
\nKidney: images and GT,
\n 1. (Hubmap): https://www.kaggle.com/competitions/hubmap-kidney-segmentation/data.
\n 2. CPTAC-CCRCC: https://wiki.cancerimagingarchive.net/display/Public/CPTAC-CCRCC
Large intestine: images
\n 1. TCGA COAD: https://wiki.cancerimagingarchive.net/display/Public/TCGA-COAD
\n 2. CPTAC-COAD: https://wiki.cancerimagingarchive.net/display/Public/CPTAC-COAD
Lung: images
\n 1. TCGA LUAD: https://wiki.cancerimagingarchive.net/display/Public/TCGA-LUAD
\n 2. TCGA LUSC: https://wiki.cancerimagingarchive.net/display/Public/TCGA-LUSC
\n 3. Lung-ACDC: https://acdc-lunghp.grand-challenge.org/Challenge/
\n 4. CPTAC-LSCC: https://wiki.cancerimagingarchive.net/display/Public/CPTAC-LSCC
\n 5. CPTAC-LUAD: https://wiki.cancerimagingarchive.net/display/Public/CPTAC-LUAD
Prostate
\n 1. TCGA-PRAD: https://wiki.cancerimagingarchive.net/display/Public/TCGA-PRAD
\n 2. PANDAs: https://www.kaggle.com/competitions/prostate-cancer-grade-assessment
\n 1. https://www.cancerimagingarchive.net/histopathology-imaging-on-tcia/
\n 2. https://wiki.cancerimagingarchive.net/display/Public/PROSTATE-MRI
Kidney + Large intestine + Lung + Spleen + Prostate:
\n (HPA) https://www.proteinatlas.org/about/download
Where are the annotations from the FTUs on the Hubmap Portal?
\nI cant find them or do they even exist?
Thanks and regards
", "votes": null, "replies": [] }, { "id": 1896346, "author_name": "jirkaborovec", "author_url": "", "post_date": "08/12/2022 19:06:13", "content": "Regarding the mentioned HuBMAP - Hacking the Kidney competition, I created a simple kernel for loading the enormous images and exporting binary masks to a dataset - https://www.kaggle.com/datasets/jirkaborovec/hacking-the-kidney-annotation-masks-png-images
\n
\n", "votes": null, "replies": [] }, { "id": 1922919, "author_name": "lililycai", "author_url": "", "post_date": "09/01/2022 20:23:31", "content": "https://www.kaggle.com/code/jirkaborovec/hubmap-segm-eda-export-rle-mask
\n
And the data here: https://cns-iu.github.io/ccf-research-kaggle-2021/
\nGtex: https://gtexportal.org/home/histologyPage
\nJust found dataset for prostate: https://www.wouterbulten.nl/blog/tech/peso-dataset-whole-slide-image-prosate-cancer/
", "votes": null, "replies": [] }, { "id": 1936994, "author_name": "sstarz", "author_url": "", "post_date": "09/13/2022 07:18:03", "content": "Very helpful! Thanks.
", "votes": null, "replies": [] }, { "id": 1969253, "author_name": "karthikpatilhub", "author_url": "", "post_date": "10/03/2022 11:57:55", "content": "is there avaliable data set
", "votes": null, "replies": [] } ], "raw_markdown_by_id": { "1836511": "## **[UPDATE]**\n\n***As per @yashvrdnjain's response, I wanted to compile all mentioned external sources here in one place. I will update the list below throughout the competition.***\n\n***If you have an external data source not listed below, please comment the details and I will add them to the table***\n\n