Do you know if there are some 2D brain datasets which are appropriate to do registration? #5
-
|
Do you know if there are some 2D brain datasets which are appropriate to do registration? |
Beta Was this translation helpful? Give feedback.
Replies: 5 comments
-
|
well, most brain imaging I know is 3D -- but we've actually just processed the OASIS dataset in a way that might help you -- after affine registration to talairach, we extracted the middle 2D slice and prepared anatomical annotations appropriate for that slice as well. I'll try to add the link and resources to this website in the next day or so. |
Beta Was this translation helpful? Give feedback.
-
|
Yes, it's standard to do affine normalization before deformable registration in 3D. I am not too familiar with the LPBA40 dataset, so I'm not sure of this second version from that description only. However, doing affine registration should be fairly straightforward on standard images/datasets. |
Beta Was this translation helpful? Give feedback.
-
|
I've added the OASIS data here, which includes a 2D (midslice) download - https://github.com/adalca/medical-datasets/blob/master/neurite-oasis.md |
Beta Was this translation helpful? Give feedback.
-
|
Yes they are affinely registered, this should all be in the readme. I can't really think of a brain 2D data set where there will be affine differences, but you could take the data set we released and apply random affine differences to the images and try to recover them.
What is the actual problem you are trying to solve? Do you have a real-world data set you are trying to solve?
…--
HMS/MIT
typed on Outlook for Android<https://aka.ms/ghei36>
________________________________
From: YMBrianHung <[email protected]>
Sent: Tuesday, February 2, 2021 11:59:57 PM
To: adalca/medical-datasets <[email protected]>
Cc: Adrian Dalca <[email protected]>; State change <[email protected]>
Subject: Re: [adalca/medical-datasets] Do you know if there are some 2D brain datasets which are appropriate to do registration? (#4)
Have the 2D dataset processed by the affine registration?
I want to train my 2D affine registration network based deep learning, Are the 2D dataset appropriate to my affine registration experiment?
--
I've added the OASIS data here, which includes a 2D (midslice) download - https://github.com/adalca/medical-datasets/blob/master/neurite-oasis.md
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub<#4 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AMBPTHEKK5A3K7DO3OC2MJ3S5CIRNANCNFSM4W6OOEJA>.
—
You are receiving this because you modified the open/close state.
Reply to this email directly, view it on GitHub<#4 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AAYSBW4BYOK63B6FKLPDQG3S5DJ43ANCNFSM4W6OOEJA>.
|
Beta Was this translation helpful? Give feedback.
-
|
Moving this to discussions |
Beta Was this translation helpful? Give feedback.
well, most brain imaging I know is 3D -- but we've actually just processed the OASIS dataset in a way that might help you -- after affine registration to talairach, we extracted the middle 2D slice and prepared anatomical annotations appropriate for that slice as well. I'll try to add the link and resources to this website in the next day or so.