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

Implementation of label map and viewing prediction.nii.gz #36

Closed
build2create opened this issue Sep 6, 2017 · 1 comment · Fixed by #25
Closed

Implementation of label map and viewing prediction.nii.gz #36

build2create opened this issue Sep 6, 2017 · 1 comment · Fixed by #25

Comments

@build2create
Copy link

build2create commented Sep 6, 2017

@ellisdg The file is produced after successfully running predict.py. But the label map (which I am assuming is prediction.nii.gz) is not multicolor. I have tried itk snap and 3D slicer. I am not able to understand how is multiclass classification is implemented. Is the tumor detected using 3D unet also segmented?If yes,does the label map give different colors to segmented portions of the tumor. For e.g. red to necrotic, blue to advancing and so on. Please suggest any other tool or alternatively can you post the screenshot of the segmented image?

Here is the screenshot of what I got using model trained with config["image_shape"]=(16,16,16). This is using itk snap:
image
And this one using 3D slicer:
image

@mystique68
Copy link

I built the model for 50 epochs but still getting the prediction.nii.gz as blank(nothing is shown up). data_T1.nii.gz is showing segmentation in red and green colors. What do these colors indicate? Do they indicate various types like necrotic,advancing regions of same tumor? Also suggest any good tool for viewing the segmented image. Can you post any sample of predicted(segmented) image?

ellisdg added a commit that referenced this issue Nov 18, 2017
1) Uses SimpleITK if N4BiasFieldCorrection cannot be found by nipype (closes #52 & closes #32).
2) Adds predict.py file that uses the trained model and writes the predicted labels to file (closes #51). The predictions are now multi-label (closes #41 & closes #36).
3) Fixes relative import (closes #49).
4) Removes "pickable" flag from training which fixes and closes #47.
5) Adds batch normalization option (closes #39).
6) Adds option to use patch training. This is will substantially reduce the memory requirement for training.
7) Updates README. Links to data that do not require registration (closes #37).
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

Successfully merging a pull request may close this issue.

2 participants