Hi,

In running the RHL6 version (and python 3.7), we are receiving consistent errors with the ImageMath and BRAINSFit tools such that the itkImageFileReader "Could not create IO object for reading file". This results in command termination and empty skullstripping, TissueSegAtlas, and VentricleMasking output folders. 

A sample error with the LargeCSF01 T1 and T2 files: Following error(s) occured : terminate called after throwing an instance of 'itk::ImmageFileReaderException' what (): /tools/ITK/ITKv4.10.0/ITKv4.10.0_THL64_stat_Release/include/ITK-4.10/itkImageFileReader.hxx:143: Could not create IO object for reading the file LargerCSF01_T1_FinalBrainMask_Erosion01.nrrd The file doesn't exist. Filename = LargeCSF01_T1_FinalBrainMask_Erosion01.nrrd

Are some dependencies missing or has something else gone awry?

Thanks,
Maddy

Hi Maddy

Do you also see "can't create IO object for writing" errors (rather than reading). If so, does the output say why it cannot write?

Since you see many of these, my possible guesses are:
- that the output directory may not have correct permissions to write in
- the disk is full
- naming problems (Are there spaces in your input filenames, any non-alphanumeric characters)

This is unlikely a dependency issue. If the above did not help, could you provide the whole terminal output as an attached file?

Martin
Originally posted by petemade:

Hi,

Thank you for your previous suggestions. I haven't seen any "can't create IO object for writing" errors--just the "reading" ones. Hence, I have attached a complete output_log.

Thanks again,
Maddy

Thanks Maddy
If you look at the output, the first step has an immidiate error (BRAINSFit). There is nothing specified for the fixedvolume (your template), which means you did not specify a template for the registration.

So, if you want to do the atlas/reference registration, you need to specify the template file in the corresponding first step tab.

Btw I see that you are using a file called ICBM_T1.nrrd. So, you probably do not need to do a reference space registration, as your data is already in ICBM space, correct? So, just skip that first step (disable that step in the first step tab).

Martin
Originally posted by petemade:

Thank you for your previous suggestions and assistance! In order to verify that our system is working, would it be possible to get an example dataset with its expected output?

Thanks again,
Maddy