search

HiLab-git / COPLE-Net

COVID-19 Pneumonia Lesion segmentation network
MIT License
87 stars 16 forks source link

own images pre-processing #6

Open Diyhhh opened 3 years ago

Diyhhh commented 3 years ago

Dear, your work is great, I have some questions and look forward to your reply. For segment COVID-19 pneumonia lesions from my images, Can you share the code for data pre-processing? you said that the images have been cropped into the lung region, and the intensity has been normalized into [0, 1] using window width/level of 1500/-650.

taigw commented 3 years ago

Hi, the normalization is quite easy. The code looks like the following:

 img[img<-1400] = -1400
 img[img> 100] = 100
 img = img + 1400
 img = img / 1500.0

For cropping, you can use manual cropping.

dscarmo commented 1 year ago

@taigw I have run your method and the results don't look good in my test set without the cropping to the lung region, but I don't have a way to reproduce your cropping. Did you use any kind of automatic cropping to the lung region? Could you provide the script in the repo? Thanks. Its not reasonable to do manual cropping in my test set since I am comparing your outputs to medical outcomes and it involves hundreds of CTs.