Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

3d dense UNET for vertebrae segmentation #9

Closed
reygz opened this issue Jan 13, 2021 · 1 comment
Closed

3d dense UNET for vertebrae segmentation #9

reygz opened this issue Jan 13, 2021 · 1 comment

Comments

@reygz
Copy link

reygz commented Jan 13, 2021

Hi, the spineweb dataset2 contains vertebrae level labels for vertebrae segmentation. Any suggestions on how to modify/extend 3d dense UNET for vertebrae segmentation?

@mrkolarik
Copy link
Owner

Hi,

well there are multiple possible routes how to proceed, but If you want to implement the vertebrae segmentation and classification as an end-to-end task using single neural network, then I would treat this problem as a multi class segmentation.

You would have to encode the data properly and then change the output layer of the net to softmax activation with as many channels as there are classes. This article describes something similar:
https://naomi-fridman.medium.com/multi-class-image-segmentation-a5cc671e647a

Hope this helps :)

Take care,

Martin

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants