As in Masks of Karolinska, We have only 0, 1 and 2. Where 2 stands for cancerous part. But it doesn't provide information about Gleason score. Then how ISUP labels are assigned for Karolinska.
", "messages": [ { "id": 848020, "postDate": "2020-05-14T17:48:44.313Z", "content": "Yes, the mask only specifies where there is cancer, not the grade. But say you have a biopsy with Gleason score 3+3, then you know that all regions containing cancer are GP 3. The same is true for 4+4 and 5+5. The mixed patterns are more challenging - but there should be some fun problems left for you to solve, right? :)
\n\nThe Gleason score and the corresponding ISUP grade is found in the csv file.
", "rawMarkdown": "Yes, the mask only specifies where there is cancer, not the grade. But say you have a biopsy with Gleason score 3+3, then you know that all regions containing cancer are GP 3. The same is true for 4+4 and 5+5. The mixed patterns are more challenging - but there should be some fun problems left for you to solve, right? :)\n\nThe Gleason score and the corresponding ISUP grade is found in the csv file.", "votes": 3, "replies": [ { "id": 848047, "postDate": "2020-05-14T17:58:33.683Z", "content": "Great. Thanks for the response. I thought I was missing something.\nThis will help. Thanks again.
", "rawMarkdown": "Great. Thanks for the response. I thought I was missing something.\nThis will help. Thanks again.", "votes": 1 } ] }, { "id": 850371, "postDate": "2020-05-16T15:07:58.673Z", "content": "Question on the Karolinska: I've noticed that for some image_id they have a data_provider of karolinska but the isup_grade is 4, which doesn't make sense, since in the data description the only labels for karolinska are 0,1 and 2. Anyone know what's wrong here? Thanks
", "rawMarkdown": "Question on the Karolinska: I've noticed that for some image_id they have a data_provider of karolinska but the isup_grade is 4, which doesn't make sense, since in the data description the only labels for karolinska are 0,1 and 2. Anyone know what's wrong here? Thanks ", "votes": 1, "replies": [ { "id": 850397, "postDate": "2020-05-16T15:31:09.853Z", "content": "actually never mind. I'll just create my own label using the Gleason score
", "rawMarkdown": "actually never mind. I'll just create my own label using the Gleason score" }, { "id": 850586, "postDate": "2020-05-16T19:41:21.147Z", "content": "Yes. Here 2 represents that tissue is cancerous or not.\n0,1 or 2 are possible values for Gleason Score not ISUP I suppose.
", "rawMarkdown": "Yes. Here 2 represents that tissue is cancerous or not.\n0,1 or 2 are possible values for Gleason Score not ISUP I suppose. " }, { "id": 851261, "postDate": "2020-05-17T13:40:54.343Z", "content": "The mask does not contain any information at all about the Gleason pattern / Gleason score / ISUP. Pixels that have value 2 in the corresponding mask is cancerous. Then you have to deduce the gleason patten in that region by considering the ISUP / Gleason score that you have for that image in the csv file.
", "rawMarkdown": "The mask does not contain any information at all about the Gleason pattern / Gleason score / ISUP. Pixels that have value 2 in the corresponding mask is cancerous. Then you have to deduce the gleason patten in that region by considering the ISUP / Gleason score that you have for that image in the csv file." } ] }, { "id": 847811, "postDate": "2020-05-14T15:58:50.403Z", "content": "As in Masks of Karolinska, We have only 0, 1 and 2. Where 2 stands for cancerous part. But it doesn't provide information about Gleason score. Then how ISUP labels are assigned for Karolinska.
", "rawMarkdown": "As in Masks of Karolinska, We have only 0, 1 and 2. Where 2 stands for cancerous part. But it doesn't provide information about Gleason score. Then how ISUP labels are assigned for Karolinska." } ], "comments": [ { "id": 848020, "author_name": "PeterStröm", "author_url": "", "post_date": "2020-05-14T17:48:44.313000", "content": "Yes, the mask only specifies where there is cancer, not the grade. But say you have a biopsy with Gleason score 3+3, then you know that all regions containing cancer are GP 3. The same is true for 4+4 and 5+5. The mixed patterns are more challenging - but there should be some fun problems left for you to solve, right? :)
\n\nThe Gleason score and the corresponding ISUP grade is found in the csv file.
", "votes": 3, "replies": [ { "id": 848047, "author_name": "Salman", "author_url": "", "post_date": "2020-05-14T17:58:33.683000", "content": "Great. Thanks for the response. I thought I was missing something.\nThis will help. Thanks again.
", "votes": 1, "replies": [] } ] }, { "id": 850371, "author_name": "Adi Mithani", "author_url": "", "post_date": "2020-05-16T15:07:58.673000", "content": "Question on the Karolinska: I've noticed that for some image_id they have a data_provider of karolinska but the isup_grade is 4, which doesn't make sense, since in the data description the only labels for karolinska are 0,1 and 2. Anyone know what's wrong here? Thanks
", "votes": 1, "replies": [ { "id": 850397, "author_name": "Adi Mithani", "author_url": "", "post_date": "2020-05-16T15:31:09.853000", "content": "actually never mind. I'll just create my own label using the Gleason score
", "votes": 0, "replies": [] }, { "id": 850586, "author_name": "Salman", "author_url": "", "post_date": "2020-05-16T19:41:21.147000", "content": "Yes. Here 2 represents that tissue is cancerous or not.\n0,1 or 2 are possible values for Gleason Score not ISUP I suppose.
", "votes": 0, "replies": [] }, { "id": 851261, "author_name": "PeterStröm", "author_url": "", "post_date": "2020-05-17T13:40:54.343000", "content": "The mask does not contain any information at all about the Gleason pattern / Gleason score / ISUP. Pixels that have value 2 in the corresponding mask is cancerous. Then you have to deduce the gleason patten in that region by considering the ISUP / Gleason score that you have for that image in the csv file.
", "votes": 0, "replies": [] } ] } ], "raw_markdown_by_id": { "848020": "Yes, the mask only specifies where there is cancer, not the grade. But say you have a biopsy with Gleason score 3+3, then you know that all regions containing cancer are GP 3. The same is true for 4+4 and 5+5. The mixed patterns are more challenging - but there should be some fun problems left for you to solve, right? :)\n\nThe Gleason score and the corresponding ISUP grade is found in the csv file.", "850371": "Question on the Karolinska: I've noticed that for some image_id they have a data_provider of karolinska but the isup_grade is 4, which doesn't make sense, since in the data description the only labels for karolinska are 0,1 and 2. Anyone know what's wrong here? Thanks ", "847811": "As in Masks of Karolinska, We have only 0, 1 and 2. Where 2 stands for cancerous part. But it doesn't provide information about Gleason score. Then how ISUP labels are assigned for Karolinska." } }