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

Captk errors I have encountered #1471

Open
MGH194 opened this issue Feb 5, 2022 · 1 comment
Open

Captk errors I have encountered #1471

MGH194 opened this issue Feb 5, 2022 · 1 comment

Comments

@MGH194
Copy link

MGH194 commented Feb 5, 2022

hi everyone
I have no experience in programming but I have been trying to work with the Captk as an end user and use it in my project. I am going to describe some problem I have encountered with when use this software.
1- for glioma segmentation:

  • load images according to this order: T1, T1CE, T2, FLAIR.
  • it couldn't work with Dicom image format for tumor segmentation application. when I load dicom images this error appear:

image001

when I try command line (you can find text file that is stored in C:\Users*your user name*.CaPTk\1.8.1 then open it and copy & past the last line command to CMD to execute this application aging) I encounter with this message:

The first image file 'C:/Users/MGH/.CaPTk/1.8.1/tmp_4669732791761504473/t1.nii.gz' has different physical dimensions with 'C:/Users/MGH/.CaPTk/1.8.1/tmp_4669732791761504473/t2.nii.gz', please register them before trying again.

so it is recommended to convert dicom image to nifti format image using preprocessing tan and select Dicom to Nifti option.

  • even in the case of loading nifti images, this message may be appear on the screen:
    image

in my case when I load T2 image this message will be appear. I don't know if it is relevant to the image sequence that we have used to obtain T2 image? in our institute we obtain T2 images using Propeller sequence (a generic name in GE MRI scanner) which is less sensitive to patient's head motion and so preserve image quality.
after press ok button a dialog box appears and you should select T2 image as moving image and T1 image as fixed image to register inconsistent T2 image with T1 image. after completing registration, load the registered image instead of inconsistent image.

  • the result of the application of brain tumor segmentation (deep learning) is strange for me. I expect to visit a folder that consist of segmented map of the tumor or at least I can visit overlaid segmented tumor on anatomical images like this picture:

image

but I visit this folders and files after completion of the application:
image

image

there is no file that show me segmented tumor.

I propose the Captk team to produce some tutorials video and instruct end user how to use their application properly.

I am trying to use this software more and write my question about it here.

thank you for your free software.

@AlexanderGetka-cbica
Copy link
Contributor

Hi @MGH194,

Thanks for your interest in CaPTk and for your feedback.

I will investigate the disparity between the CLI and GUI modes of the DeepMedic-based brain tumor segmentation.

Regarding loading T2 and T1 images together:
This error is shown when images have inconsistent dimensions with one another and can't be loaded simultaneously in the viewer. As you mentioned, registering the two images allows you to load them together -- I would guess that the T2 scanner is producing images with a different resolution or origin. You can check this from the command line with our Utilities tool, like so:

CaPTk Utilities -inf -i C:/path/to/image/file.nii.gz

Regarding the output of the brain tumor segmentation application:
I believe the Segm.nii file contains the multi-label segmentation. From the CaPTk GUI, first load the registered image file you used as input, then go to File -> Load -> ROI and load the Segm.nii file that way. This should give you the visualization you want.

We have tutorials on our documentation at https://cbica.github.io/CaPTk/How_To_Guides.html , as well as a few video tutorials on this YouTube channel: https://www.youtube.com/channel/UC69N7TN5bH2onj4dHcPLxxA

Thanks again for reaching out and I hope this solves your issue. Please feel free to let us know if any of the above is unclear or if you have additional feedback.

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