(EDIT): Given the interest I changed the title to a more generic thread for anyone wanting to contribute. If you leave comments I can add them to this top Topic for easy access.
\n\n(EDIT-2): @tpmeli has also collected some amazing resources on his post that wonderfully augment what's here!
\n\nI wanted to contribute a short Literature Review on previous methods and approaches in order to help jump start the group. If this is useful I am happy to expand on either depth or breadth of the search just let me know.
\n\nDeep learning for automatic Gleason pattern classification for grade group determination of prostate biopsies\na. Good into to the problem set and descriptions of pre-processing\nb. Sets baselines for other work as well as benchmarks of performance
Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis\na. Longer and more generic for CNN's use across various cancers\nb. @geertlitjens is a contributing author and also working this Kaggle!
Comparison of Artificial Intelligence Techniques to Evaluate Performance of a Classifier for Automatic Grading of Prostate Cancer From Digitized Histopathologic Images\na. Direct comparison on methodologies.\nb. Results:On 333 tissue microarray cores from 231 participants with prostate cancer (mean [SD] age, 63.2 [6.3] years), 20-fold leave-patches-out CV resulted in mean (SD) accuracy of 97.8% (1.2%), sensitivity of 98.5% (1.0%), and specificity of 97.5% (1.2%) for classifying benign patches vs cancerous patches.
Automated Gleason grading of prostate cancer tissue microarrays via deep learning\na. Nice Github Repo -> gleson_CNN
@wouterbulten - From the contributors as paper with more information on the background of the data and problems we encountered: \na. Ström & Kartasalo et al., 2020. Artificial intelligence for diagnosis and grading of prostate cancer in biopsies: a population-based, diagnostic study. The Lancet Oncology\n-Full Text\nb. Bulten et al., 2020. Automated deep-learning system for Gleason grading of prostate cancer using biopsies: a diagnostic study. The Lancet Oncology.\n-Full Text
(EDIT): Given the interest I changed the title to a more generic thread for anyone wanting to contribute. If you leave comments I can add them to this top Topic for easy access.
\n\n(EDIT-2): @tpmeli has also collected some amazing resources on his post that wonderfully augment what's here!
\n\nI wanted to contribute a short Literature Review on previous methods and approaches in order to help jump start the group. If this is useful I am happy to expand on either depth or breadth of the search just let me know.
\n\nDeep learning for automatic Gleason pattern classification for grade group determination of prostate biopsies\na. Good into to the problem set and descriptions of pre-processing\nb. Sets baselines for other work as well as benchmarks of performance
Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis\na. Longer and more generic for CNN's use across various cancers\nb. @geertlitjens is a contributing author and also working this Kaggle!
Comparison of Artificial Intelligence Techniques to Evaluate Performance of a Classifier for Automatic Grading of Prostate Cancer From Digitized Histopathologic Images\na. Direct comparison on methodologies.\nb. Results:On 333 tissue microarray cores from 231 participants with prostate cancer (mean [SD] age, 63.2 [6.3] years), 20-fold leave-patches-out CV resulted in mean (SD) accuracy of 97.8% (1.2%), sensitivity of 98.5% (1.0%), and specificity of 97.5% (1.2%) for classifying benign patches vs cancerous patches.
Automated Gleason grading of prostate cancer tissue microarrays via deep learning\na. Nice Github Repo -> gleson_CNN
@wouterbulten - From the contributors as paper with more information on the background of the data and problems we encountered: \na. Ström & Kartasalo et al., 2020. Artificial intelligence for diagnosis and grading of prostate cancer in biopsies: a population-based, diagnostic study. The Lancet Oncology\n-Full Text\nb. Bulten et al., 2020. Automated deep-learning system for Gleason grading of prostate cancer using biopsies: a diagnostic study. The Lancet Oncology.\n-Full Text
Completely awesome. I linked to you in a similar post I did. Thanks for sharing!
", "rawMarkdown": "Completely awesome. I linked to you in a [similar post](https://www.kaggle.com/c/prostate-cancer-grade-assessment/discussion/145682) I did. Thanks for sharing!", "votes": 1, "replies": [ { "id": 819614, "postDate": "2020-04-24T17:57:01.833Z", "content": "Awesome, to I'll do the same to make sure no one misses out if they only come across one 👍
", "rawMarkdown": "Awesome, to I'll do the same to make sure no one misses out if they only come across one 👍 ", "votes": 1 } ] }, { "id": 816636, "postDate": "2020-04-22T13:34:46.930Z", "content": "Nice overview! If you want to do a deep-dive and are interested in the work we have been doing before this competition, I can recommend these two articles:
\n\nThese papers also contain more information on the background of the data and problems we encountered.
", "rawMarkdown": "Nice overview! If you want to do a deep-dive and are interested in the work we have been doing before this competition, I can recommend these two articles:\n\n- Ström & Kartasalo et al., 2020. Artificial intelligence for diagnosis and grading of prostate cancer in biopsies: a population-based, diagnostic study. The Lancet Oncology. (freely accessible preprint here: https://arxiv.org/abs/1907.01368)\nhttps://www.thelancet.com/journals/lanonc/article/PIIS1470-2045(19)30738-7/fulltext\n- Bulten et al., 2020. Automated deep-learning system for Gleason grading of prostate cancer using biopsies: a diagnostic study. The Lancet Oncology. (freely accesible preprint here: https://arxiv.org/abs/1907.07980)\nhttps://www.thelancet.com/journals/lanonc/article/PIIS1470-2045(19)30739-9/fulltext\n\nThese papers also contain more information on the background of the data and problems we encountered.", "votes": 2, "replies": [ { "id": 816673, "postDate": "2020-04-22T14:01:04.720Z", "content": "Awesome thanks! I edited the post some as I think it can just serve as a ongoing thread.
", "rawMarkdown": "Awesome thanks! I edited the post some as I think it can just serve as a ongoing thread. ", "votes": 2 }, { "id": 816680, "postDate": "2020-04-22T14:06:43.620Z", "content": "Added these papers to the top thread!
", "rawMarkdown": "Added these papers to the top thread!", "votes": 1 } ] }, { "id": 816573, "postDate": "2020-04-22T12:40:17.857Z", "content": "Thanks for sharing.
", "rawMarkdown": "Thanks for sharing.", "votes": 1 } ], "comments": [ { "id": 819485, "author_name": "Tom M", "author_url": "", "post_date": "2020-04-24T16:13:53.357000", "content": "Completely awesome. I linked to you in a similar post I did. Thanks for sharing!
", "votes": 1, "replies": [ { "id": 819614, "author_name": "Zac Dannelly", "author_url": "", "post_date": "2020-04-24T17:57:01.833000", "content": "Awesome, to I'll do the same to make sure no one misses out if they only come across one 👍
", "votes": 1, "replies": [] } ] }, { "id": 816636, "author_name": "Wouter Bulten", "author_url": "", "post_date": "2020-04-22T13:34:46.930000", "content": "Nice overview! If you want to do a deep-dive and are interested in the work we have been doing before this competition, I can recommend these two articles:
\n\nThese papers also contain more information on the background of the data and problems we encountered.
", "votes": 2, "replies": [ { "id": 816673, "author_name": "Zac Dannelly", "author_url": "", "post_date": "2020-04-22T14:01:04.720000", "content": "Awesome thanks! I edited the post some as I think it can just serve as a ongoing thread.
", "votes": 2, "replies": [] }, { "id": 816680, "author_name": "Zac Dannelly", "author_url": "", "post_date": "2020-04-22T14:06:43.620000", "content": "Added these papers to the top thread!
", "votes": 1, "replies": [] } ] }, { "id": 816573, "author_name": "Dracarys", "author_url": "", "post_date": "2020-04-22T12:40:17.857000", "content": "Thanks for sharing.
", "votes": 1, "replies": [] } ], "raw_markdown_by_id": { "816547": "(EDIT): Given the interest I changed the title to a more generic thread for anyone wanting to contribute. If you leave comments I can add them to this top Topic for easy access. \n\n(EDIT-2): @tpmeli has also collected some [amazing resources on his post](https://www.kaggle.com/c/prostate-cancer-grade-assessment/discussion/145682) that wonderfully augment what's here!\n\nI wanted to contribute a short Literature Review on previous methods and approaches in order to help jump start the group. If this is useful I am happy to expand on either depth or breadth of the search just let me know. \n\n1. [Deep learning for automatic Gleason pattern classification for grade group determination of prostate biopsies](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6611751/)\na. Good into to the problem set and descriptions of pre-processing\nb. Sets baselines for other work as well as benchmarks of performance \n\n2. [Deep learning as a tool for increased accuracy and efficiency of histopathological diagnosis](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4876324/)\na. Longer and more generic for CNN's use across various cancers\nb. @geertlitjens is a contributing author and also working this Kaggle!\n\n3. [Comparison of Artificial Intelligence Techniques to Evaluate Performance of a Classifier for Automatic Grading of Prostate Cancer From Digitized Histopathologic Images](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6484626/)\na. Direct comparison on methodologies.\nb. Results:On 333 tissue microarray cores from 231 participants with prostate cancer (mean [SD] age, 63.2 [6.3] years), 20-fold leave-patches-out CV resulted in mean (SD) accuracy of 97.8% (1.2%), sensitivity of 98.5% (1.0%), and specificity of 97.5% (1.2%) for classifying benign patches vs cancerous patches. \n\n4. [Automated Gleason grading of prostate cancer tissue microarrays via deep learning](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6089889/)\na. Nice Github Repo -> [gleson_CNN](https://github.com/eiriniar/gleason_CNN)\n\n5. @wouterbulten - From the contributors as paper with more information on the background of the data and problems we encountered: \na. [Ström & Kartasalo et al., 2020. Artificial intelligence for diagnosis and grading of prostate cancer in biopsies: a population-based, diagnostic study. The Lancet Oncology](https://arxiv.org/abs/1907.01368)\n -[Full Text](https://www.thelancet.com/journals/lanonc/article/PIIS1470-2045(19)30738-7/fulltext)\nb. [Bulten et al., 2020. Automated deep-learning system for Gleason grading of prostate cancer using biopsies: a diagnostic study. The Lancet Oncology.](https://arxiv.org/abs/1907.07980)\n -[Full Text](https://www.thelancet.com/journals/lanonc/article/PIIS1470-2045(19)30739-9/fulltext)", "819485": "Completely awesome. I linked to you in a [similar post](https://www.kaggle.com/c/prostate-cancer-grade-assessment/discussion/145682) I did. Thanks for sharing!", "816636": "Nice overview! If you want to do a deep-dive and are interested in the work we have been doing before this competition, I can recommend these two articles:\n\n- Ström & Kartasalo et al., 2020. Artificial intelligence for diagnosis and grading of prostate cancer in biopsies: a population-based, diagnostic study. The Lancet Oncology. (freely accessible preprint here: https://arxiv.org/abs/1907.01368)\nhttps://www.thelancet.com/journals/lanonc/article/PIIS1470-2045(19)30738-7/fulltext\n- Bulten et al., 2020. Automated deep-learning system for Gleason grading of prostate cancer using biopsies: a diagnostic study. The Lancet Oncology. (freely accesible preprint here: https://arxiv.org/abs/1907.07980)\nhttps://www.thelancet.com/journals/lanonc/article/PIIS1470-2045(19)30739-9/fulltext\n\nThese papers also contain more information on the background of the data and problems we encountered.", "816573": "Thanks for sharing." } }