The title basically explains it all. What values in the metadata are relevant to make a prediction. It seems that a lot of those repeat quite a lot and there are others such as TableHeight that I'm not sure can help us make a prediction and some that I don't know what they represent at all (like KVP). Can anyone help me out on this?
", "messages": [ { "id": "986889", "postDate": "08/26/2020 21:20:18", "content": "The title basically explains it all. What values in the metadata are relevant to make a prediction. It seems that a lot of those repeat quite a lot and there are others such as TableHeight that I'm not sure can help us make a prediction and some that I don't know what they represent at all (like KVP). Can anyone help me out on this?
", "rawMarkdown": "The title basically explains it all. What values in the metadata are relevant to make a prediction. It seems that a lot of those repeat quite a lot and there are others such as TableHeight that I'm not sure can help us make a prediction and some that I don't know what they represent at all (like KVP). Can anyone help me out on this?", "votes": null }, { "id": "987190", "postDate": "08/27/2020 04:39:24", "content": "I tried using the DICOM metadata from prediction, but couldn't really find anything that is specific to the patient themselves. So, I don't think there is much in the metadata that's gonna be helpful in the prediction.
", "rawMarkdown": "I tried using the DICOM metadata from prediction, but couldn't really find anything that is specific to the patient themselves. So, I don't think there is much in the metadata that's gonna be helpful in the prediction.", "votes": null }, { "id": "987469", "postDate": "08/27/2020 09:43:12", "content": "how about the slice location? that can differ from one patient to another. And do you happen to know what KVP means and is it useful?
", "rawMarkdown": "how about the slice location? that can differ from one patient to another. And do you happen to know what KVP means and is it useful?", "votes": null }, { "id": "987726", "postDate": "08/27/2020 13:23:27", "content": "I might be completely wrong here, but with a quick google search it looks like KVP means kilovoltage peak of the scanner and I don't think it would be related to the FVP of the patient. and the same for the slice location. Do correct me if I'm wrong.
", "rawMarkdown": "I might be completely wrong here, but with a quick google search it looks like KVP means kilovoltage peak of the scanner and I don't think it would be related to the FVP of the patient. and the same for the slice location. Do correct me if I'm wrong.", "votes": null }, { "id": "987806", "postDate": "08/27/2020 14:33:05", "content": "I think Slice Location might have some useful information. Including Slice Location will allow the network to compare lungs taken at the same location. For example, assume two patients had a slice at the same location, but one patient had the slice at the bottom of the lung, and the other from about the middle. This will allow the network to infer the size of the lung.
\nOr even better there is Image Location which has coordinates in 3D space. But I'm not sure about the other data.
Hi @darkcube
\ncould u help understand the terms
\n1) slice location
\n2) slicespacings
\n3) ImageLocation
\nHow can we interpret them as
Image location is just the 3D coordinates of where the image was taken. Slice location is the z component of that. Slice spacing is distance between the slice and the other.
", "rawMarkdown": "Image location is just the 3D coordinates of where the image was taken. Slice location is the z component of that. Slice spacing is distance between the slice and the other.", "votes": null }, { "id": "990768", "postDate": "08/29/2020 20:33:05", "content": "https://dicom.innolitics.com/ciods/ct-image/image-plane/00201041
\nThis website contains information about a lot of the DICOM attributes, if you didn't already know.
Oh, thank you so much. I didn't know.
", "rawMarkdown": "Oh, thank you so much. I didn't know.", "votes": null }, { "id": "990964", "postDate": "08/30/2020 02:20:32", "content": "I did go through that earlier… but it was not so layman friendly description.so was hoping for simplified explanation
", "rawMarkdown": "I did go through that earlier... but it was not so layman friendly description.so was hoping for simplified explanation", "votes": null }, { "id": "992089", "postDate": "08/30/2020 22:46:47", "content": "personaly i used them all exepte unique value features.
", "rawMarkdown": "personaly i used them all exepte unique value features.", "votes": null }, { "id": "992102", "postDate": "08/30/2020 23:30:51", "content": "you're Algerian! we're from the same country!
", "rawMarkdown": "you're Algerian! we're from the same country!", "votes": null }, { "id": "992103", "postDate": "08/30/2020 23:32:32", "content": "this is the first time I find someone from my country here.
\nالسلام عليكم
", "rawMarkdown": "this is the first time I find someone from my country here. \n\nالسلام عليكم", "votes": null } ], "comments": [ { "id": 987190, "author_name": "jonykarki", "author_url": "", "post_date": "08/27/2020 04:39:24", "content": "I tried using the DICOM metadata from prediction, but couldn't really find anything that is specific to the patient themselves. So, I don't think there is much in the metadata that's gonna be helpful in the prediction.
", "votes": null, "replies": [ { "id": 987469, "author_name": "darkcube", "author_url": "", "post_date": "08/27/2020 09:43:12", "content": "how about the slice location? that can differ from one patient to another. And do you happen to know what KVP means and is it useful?
", "votes": null, "replies": [] }, { "id": 987726, "author_name": "jonykarki", "author_url": "", "post_date": "08/27/2020 13:23:27", "content": "I might be completely wrong here, but with a quick google search it looks like KVP means kilovoltage peak of the scanner and I don't think it would be related to the FVP of the patient. and the same for the slice location. Do correct me if I'm wrong.
", "votes": null, "replies": [] }, { "id": 987806, "author_name": "darkcube", "author_url": "", "post_date": "08/27/2020 14:33:05", "content": "I think Slice Location might have some useful information. Including Slice Location will allow the network to compare lungs taken at the same location. For example, assume two patients had a slice at the same location, but one patient had the slice at the bottom of the lung, and the other from about the middle. This will allow the network to infer the size of the lung.
\nOr even better there is Image Location which has coordinates in 3D space. But I'm not sure about the other data.
Hi @darkcube
\ncould u help understand the terms
\n1) slice location
\n2) slicespacings
\n3) ImageLocation
\nHow can we interpret them as
Image location is just the 3D coordinates of where the image was taken. Slice location is the z component of that. Slice spacing is distance between the slice and the other.
", "votes": null, "replies": [] }, { "id": 990768, "author_name": "jonykarki", "author_url": "", "post_date": "08/29/2020 20:33:05", "content": "https://dicom.innolitics.com/ciods/ct-image/image-plane/00201041
\nThis website contains information about a lot of the DICOM attributes, if you didn't already know.
Oh, thank you so much. I didn't know.
", "votes": null, "replies": [] }, { "id": 990964, "author_name": "jaideepvalani", "author_url": "", "post_date": "08/30/2020 02:20:32", "content": "I did go through that earlier… but it was not so layman friendly description.so was hoping for simplified explanation
", "votes": null, "replies": [] } ] }, { "id": 992089, "author_name": "servietsky", "author_url": "", "post_date": "08/30/2020 22:46:47", "content": "personaly i used them all exepte unique value features.
", "votes": null, "replies": [ { "id": 992102, "author_name": "darkcube", "author_url": "", "post_date": "08/30/2020 23:30:51", "content": "you're Algerian! we're from the same country!
", "votes": null, "replies": [] }, { "id": 992103, "author_name": "darkcube", "author_url": "", "post_date": "08/30/2020 23:32:32", "content": "this is the first time I find someone from my country here.
\nالسلام عليكم
", "votes": null, "replies": [] } ] } ], "raw_markdown_by_id": { "986889": "The title basically explains it all. What values in the metadata are relevant to make a prediction. It seems that a lot of those repeat quite a lot and there are others such as TableHeight that I'm not sure can help us make a prediction and some that I don't know what they represent at all (like KVP). Can anyone help me out on this?", "987190": "I tried using the DICOM metadata from prediction, but couldn't really find anything that is specific to the patient themselves. So, I don't think there is much in the metadata that's gonna be helpful in the prediction.", "987469": "how about the slice location? that can differ from one patient to another. And do you happen to know what KVP means and is it useful?", "987726": "I might be completely wrong here, but with a quick google search it looks like KVP means kilovoltage peak of the scanner and I don't think it would be related to the FVP of the patient. and the same for the slice location. Do correct me if I'm wrong.", "987806": "I think Slice Location might have some useful information. Including Slice Location will allow the network to compare lungs taken at the same location. For example, assume two patients had a slice at the same location, but one patient had the slice at the bottom of the lung, and the other from about the middle. This will allow the network to infer the *size* of the lung. \nOr even better there is *Image* Location which has coordinates in 3D space. But I'm not sure about the other data.", "988608": "Hi @darkcube \ncould u help understand the terms\n1) slice location\n2) slicespacings \n3) ImageLocation\nHow can we interpret them as", "990724": "Image location is just the 3D coordinates of where the image was taken. Slice location is the z component of that. Slice spacing is distance between the slice and the other.", "990768": "[https://dicom.innolitics.com/ciods/ct-image/image-plane/00201041](https://dicom.innolitics.com/ciods/ct-image/image-plane/00201041)\nThis website contains information about a lot of the DICOM attributes, if you didn't already know.", "990779": "Oh, thank you so much. I didn't know.", "990964": "I did go through that earlier... but it was not so layman friendly description.so was hoping for simplified explanation", "992089": "personaly i used them all exepte unique value features.", "992102": "you're Algerian! we're from the same country!", "992103": "this is the first time I find someone from my country here. \n\nالسلام عليكم" }, "source": "meta" }