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

Image #4

Open
saranggoel opened this issue Nov 28, 2020 · 7 comments
Open

Image #4

saranggoel opened this issue Nov 28, 2020 · 7 comments

Comments

@saranggoel
Copy link

Hello,

Thank you for uploading such a great dataset! I was opening some of the images up, and for some reason, they look like this:

image

Additionally, when I try to use other .tif images with your code (the one for COVID-19 classification), it doesn't work. They look like this:

image

Do you know what may be the problem? I think you have a different format or something for your tif files versus mine, but I am not sure.

@mr7495
Copy link
Owner

mr7495 commented Nov 28, 2020

Hi;
Thank you for your interest in our work!
The original Images in our dataset are in TIFF format (16-bit uint type), which contain the same image data as the DICOM format and only do not include the patients metadata. But you can not visualize this TIFF format images with your monitor and they look like completely black images (because standard monitors can not show 16-bit images in uint type). So if you would like to convert these images to visualizable images you can use [This code.](https://github.com/mr7495/COVID-CTset/blob/master/Visualize.py)
My suggestion is to use the original images for training or testing the models in order to not lose any data.
Some of the images converted to visualizable format are:
vis_data.zip

@saranggoel
Copy link
Author

Thank you for your prompt reply! I realized why my other tif images do not work with the code (as they are in float32 format, rather than uint16 format). Do you know a way to convert these images into the uint16 type?

@saranggoel
Copy link
Author

Additionally, if I was to use the visualizer code, how can I convert the result back to the original image?

@latexers
Copy link

hi, thank you for your so excited work, have you tried convert these images to nii.gz format? that is commonly in medical image analysis, do you have some experience about this?

@saranggoel-nr
Copy link

Hey, @latexers ya I asked that question a pretty long time back and I eventually figured out how to convert it to nii.gz, but thanks for your help!

@mr7495
Copy link
Owner

mr7495 commented Jun 21, 2021

Hello all, No, unfortunately, we did not convert the images to 3D NII format, but I think the code for converting the images to this format can be access through other websites.

@latexers
Copy link

latexers commented Jul 6, 2021

Hey, @latexers ya I asked that question a pretty long time back and I eventually figured out how to convert it to nii.gz, but thanks for your help!

I found a tool for this convter: https://github.com/rordenlab/dcm2niix

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

4 participants