Home

Welcome to the sHDR wiki!

On this wiki you will find information on using HDR MRI algorithms and programs for your data.

Step - Coregistration

The images to be used to create the fused HDR image needs to be all in the same space. Pixel correspondence is used to find the best signal among the individual images and this requires that the images have pixel wise correspondence. This can be achieved using registration.

Step - Normalised intensity range

The images must have the same intensity ranges to be treated fairly unless you know what you're doing. The GUI program automatically does this but the command line application does not. Use the test app itkHighDynamicRangeImagePreprocessingTest to do the normalising.

itkHighDynamicRangeImagePreprocessingTest.exe .\T1FLIP80_to_T2MR.nii.gz 0 .\preproc\T1FLIP80_to_T2MR.nii.gz

Step - Apply HDR algorithm

You can now run the HDR algorithm on the processed data. The sHDR GUI application has a wizard that you can use. An example usage for the command line is

itkHighDynamicRangeImageApp.exe --average -m --weight 1.0 --lambda 0.8 -b 0.8 -r 0.04 -d 4 -l 3 -p output\HDR_006 .\T2MR.nii.gz .\T1FLIP80_to_T2MR.nii.gz .\T2STAR_to_T2MR_aniso.nii.gz

which also outputs the average image for comparison. Note that the sigma range and domain will need to match the intensity range of the images.