Dear all,
\n\nI was wondering, from a medical perspective, if someone is classified in category 3, they will still show the symptoms of category 1 and 2 on the retina image? Or do the early signs disappear/get replaced by the new symptoms that would put someone if disease category 3?
\n\nThank you!
", "messages": [ { "id": 596500, "postDate": "2019-08-10T18:18:43.280Z", "content": "Dear all,
\n\nI was wondering, from a medical perspective, if someone is classified in category 3, they will still show the symptoms of category 1 and 2 on the retina image? Or do the early signs disappear/get replaced by the new symptoms that would put someone if disease category 3?
\n\nThank you!
", "rawMarkdown": "Dear all,\n\nI was wondering, from a medical perspective, if someone is classified in category 3, they will still show the symptoms of category 1 and 2 on the retina image? Or do the early signs disappear/get replaced by the new symptoms that would put someone if disease category 3?\n\nThank you!", "votes": 4 }, { "id": 598135, "postDate": "2019-08-13T07:25:39.667Z", "content": "There are various grading protocols for diabetic retinopathy. For example: Messidor dataset is graded using following:
\n\n\n\n\n0 -> (Normal): (μA = 0) AND (H = 0)\n 1 -> (0 < μA < = 5) AND (H = 0)
\n
\n 2 -> ((5 < μA < 15) OR (0 < H < 5)) AND (NV = 0)
\n 3 ->(μA >= 15) OR (H >=5) OR (NV = 1)\n where:\n μA: number of microaneurysms\n H: number of hemorrhages\n NV = 1: neovascularization\n NV = 0: no neovascularization
There are others like Scottish grading protocol.\nDepending on above, category 2 will have symptoms of category 1. For example: If patient has 7 μA and 1 H (category 2), he still has minimum required for category 1.\nSo I guess early signs do not disappear, but is enhanced in a way.
\n\nNote: This is my own understanding. I'll really appreciate if someone with domain knowledge can shed more light.
", "rawMarkdown": "There are various grading protocols for diabetic retinopathy. For example: [Messidor dataset](http://www.adcis.net/en/third-party/messidor/) is graded using following:\n> \n0 -> (Normal): (μA = 0) AND (H = 0)\n1 -> (0 < μA < = 5) AND (H = 0)Do you know how current competition's dataset is graded? or how is its grading different than Messidor's?
", "rawMarkdown": "Do you know how current competition's dataset is graded? or how is its grading different than Messidor's?" }, { "id": 598977, "postDate": "2019-08-14T10:11:00.870Z", "content": "Not sure, but I think they may be using \"International clinical diabetic retinopathy disease severity scale\", because it's metioned in this pdf
", "rawMarkdown": "Not sure, but I think they may be using \"*International clinical diabetic retinopathy disease severity scale*\", because it's metioned in [this pdf](https://www.kaggle.com/c/aptos2019-blindness-detection/discussion/103854#latest-597549)" } ] }, { "id": 599437, "postDate": "2019-08-15T00:50:26.803Z", "rawMarkdown": "", "isDeleted": true } ], "comments": [ { "id": 598135, "author_name": "Ashwani Pandey", "author_url": "", "post_date": "2019-08-13T07:25:39.667000", "content": "There are various grading protocols for diabetic retinopathy. For example: Messidor dataset is graded using following:
\n\n\n\n\n0 -> (Normal): (μA = 0) AND (H = 0)\n 1 -> (0 < μA < = 5) AND (H = 0)
\n
\n 2 -> ((5 < μA < 15) OR (0 < H < 5)) AND (NV = 0)
\n 3 ->(μA >= 15) OR (H >=5) OR (NV = 1)\n where:\n μA: number of microaneurysms\n H: number of hemorrhages\n NV = 1: neovascularization\n NV = 0: no neovascularization
There are others like Scottish grading protocol.\nDepending on above, category 2 will have symptoms of category 1. For example: If patient has 7 μA and 1 H (category 2), he still has minimum required for category 1.\nSo I guess early signs do not disappear, but is enhanced in a way.
\n\nNote: This is my own understanding. I'll really appreciate if someone with domain knowledge can shed more light.
", "votes": 2, "replies": [ { "id": 598911, "author_name": "Rishabh Agrahari", "author_url": "", "post_date": "2019-08-14T08:04:01.130000", "content": "Do you know how current competition's dataset is graded? or how is its grading different than Messidor's?
", "votes": 0, "replies": [] }, { "id": 598977, "author_name": "Ashwani Pandey", "author_url": "", "post_date": "2019-08-14T10:11:00.870000", "content": "Not sure, but I think they may be using \"International clinical diabetic retinopathy disease severity scale\", because it's metioned in this pdf
", "votes": 0, "replies": [] } ] }, { "id": 599437, "author_name": "", "author_url": "", "post_date": "2019-08-15T00:50:26.803000", "content": "", "votes": 0, "replies": [] } ], "raw_markdown_by_id": { "596500": "Dear all,\n\nI was wondering, from a medical perspective, if someone is classified in category 3, they will still show the symptoms of category 1 and 2 on the retina image? Or do the early signs disappear/get replaced by the new symptoms that would put someone if disease category 3?\n\nThank you!", "598135": "There are various grading protocols for diabetic retinopathy. For example: [Messidor dataset](http://www.adcis.net/en/third-party/messidor/) is graded using following:\n> \n0 -> (Normal): (μA = 0) AND (H = 0)\n1 -> (0 < μA < = 5) AND (H = 0)