study/subject not appearing in database

[rvboutin]

Hi,
I am using BrainVisa 4.1.1 under Linux Ubuntu 12.04 LTS 64bit.
I have a "subject" called "nema_2013_05_07" in the study folder "NEMA_phantom" and for a strange reason neither the study nor the subject appears in the BrainVisa database. I tried to move the "nema_2013_05_07" subject folder into another study folder renamed "phantom" containing other files, all the other files appear correctly in the database but the NEMA one is still not. I have been scratching my head about it for the past 2 hours and cannot find any explanation....
There are no space or special characters in the folders name nor in the file names (except a "." but this is also the case for another phantom and it works fine).
Any ideas?
Thanks
Rv

[riviere]

Hi,
The study/subject names seem OK.
Have you updated the database ?
Denis

[rvboutin]

Hi,
Yes I did update the databases. I even tried to delete all the database files and update again. I also tried to copy just these folder on another disk and create a new database, it still does not work...
Cheers,
Rv

[riviere]

Then I guess something else is wrong in the files organization. How have the data of this subject been included in the brainvisa database ? Is there a T1 MRI, and has is been imported using the import process ?
The raw T1 MRI should be located like this:

Code: Select all

<database_dir>/NEMA_phantom/nema_2013_05_07/t1mri/default_acquisition/nema_2013_05_07.nii

Here "default_acquisition" may be a different directory, but the MRI file should have the same file name as the subject directory. The MRI file extension may be different if it is a different format (.ima+.dim, .nii.gz, .mnc, .img+.hdr etc).
You can also check things using the "database explorer", a process in "data management". It shows database directories trees like a file system browser, but shows files/directories which are recognized within BV databases or not: when clicking on a data or a directory, the right panel displays information like data type and database organization properties.
BrainVisa 4.1 is rather old but it should work...
Denis

[rvboutin]

Hi,
OK so, I checked the name of the files, there was a discrepancy, i.e. the folder name was in lower case and the file name in upper case... I changed this, no effect.
Just to clarify, I have only PET and CT files in each subject folders, there is no T1 MRI associated with this study. To avoid mistake when I copy files onto the hard disk, I use a template of folder and subfolder so that they are all identical.
I used the database browser as you suggested, and I can see the NEMA_2013_05_07 folder (subject) in the "Phantom" study. However, I am not too sure about what I should see or look for in the right panel of the database browser?
I have attached a screenshot of the database browser view for the Phantom study folder and sub-folders. All folders have a tag "identified=True" whereas all the pet.v files have the their tag "identified=False". Am I right in guessing that this is the culprit? If so, what's wrong? How to make those files identified by BV?
Cheers,
Rv

[riviere]

Yes, the "identified=False" is the problem (or the "?" on the icon on the left panel of the database explorer).
Was the nuclear imaging toolbox in the standard brainvisa 4.1 ? I don't remember... and I guess it is not included in later releases (which perhaps explains why you are using version 4.1)
However, I had a look at the files organization: PET images have a prefix to distiguish betxween human/animal and type of camera:
humans:
- h* for HR+1 camera
- i* for HR+2 camera
- j* for HRRT camera
- T* or PET_* for Biograph
phantoms:
- f* for HR+1
- g* for HR+2
- k* for HRRT
- m* for FOCUS
animals:
- a* for HR+1
- b* for HR+2
- c* for HRRT
- u0* for FOCUS
- v0* for INVEON
Don't ask me why those prefixes, ar what they change in processings, I'm not a PET expert.
Anyway that explains why "brain_phantom..." is recognized: starting with "b" it is recognized as an animal, HR+2 camera PET dataset, but "NEMA..." is not, because it does not start with one of the listed prefixes.
So you should prepend the correct prefix in the filename.

Denis

[rvboutin]

Hi Denis,
Spot on! That was the culprit!! Changing the first characters of the PET file to "m_" solved the problem, I can now select the files in Brainvisa.
Thanks a lot!
Rv