Hello,
I have a question regarding the submission tutorial for the "Synthesis (Global) - Missing MRI" challenge.
For the official calculation of SSIM, should the script provided at [this link](https://github.com/hongweilibran/BraSyn/blob/main/ssim.py) be used?
If so, could you please confirm whether it is correct to multiply the clipped prediction values by two as shown in line 66 of the script (prediction = prediction*2)?
Thank you very much for your assistance.
Best regards,
Created by Hyunjun Cho chohj1111 Hi, @ShadowTwin41 The second one is correct, according to last year's experience. Actually, you can use medpref to check whether or not your output format is correct. Dear all,
For the testing phase, I understand that we have to generate the missing modality and save the generated modality in the same naming and format as the original dataset.
However, I didn't understand completely if it is supposed for our code also copy the files used in the input folder, e.g. what option is correct if the missing modality is T2W?
```
Option 1:
| output_path
| | BraTS-GLI-00000-000
| | | BraTS-GLI-00000-000-t1c.nii.gz
| | | BraTS-GLI-00000-000-t1n.nii.gz
| | | BraTS-GLI-00000-000-t2f.nii.gz
| | | BraTS-GLI-00000-000-t2w.nii.gz
Option 2:
| output_path
| | BraTS-GLI-00000-000
| | | BraTS-GLI-00000-000-t2w.nii.gz
```
Also, should the data be normalized between -1 and 1? As we don't have access to the ground truth, we don't know between what values should the scan be, am I correct?
Thank you very much!
@branhongweili Thanks! Hi @chohj1111
Thank you for spotting this. It is a mistake.
We have corrected it.
Best regards,
Bran
Drop files to upload
Inquiry about the evaluation code in BraSyn challenge (Missing MRI) page is loading…