Print

Print


Hi Bennet,

That may help, if possible. What would also help is if you could prepare and send to me a minimal data set which I can use to reproduce the problem. I've not been able to reproduce it yet.

Thanks,

Paul

On Jan 31 2020, at 12:07 pm, Bennet Fauber <[log in to unmask]> wrote:
Paul,

No, not that I noticed. We're starting it by double-clicking the
FSLeyes.app folder. Do you want us to try to start it from a
Terminal?

-- bennet

On Thu, Jan 30, 2020 at 5:46 AM paul mccarthy <[log in to unmask]> wrote:

Hi Bennet,

This is a bug - is anything printed to the terminal when the error occurs?

Paul

On Jan 29 2020, at 11:42 pm, Bennet Fauber <[log in to unmask]> wrote:

I have a user who downloaded and installed FSLeyes 0.32.0 for Mac,
.zip file. She is in a workshop, and they are working with the
standard ds102 dataset from OpenfMRI.org (OpenNeuro).

She uses FSL to create from sub-08_T1w.nii.gz using bet the mask file,
sub-08_T1w_brain.nii.gz.

When she uses FSLeyes 0.32.0 to open first the T1w file, that works,
but when she tries to add the mask as an overlay she gets the error
window attached, which says, basically, that the mask file isn't
proper BIDS. In case the screen shot does not pass the listserver,
the text is roughly,

Error loading overlay
An error occurred loading the image /path/to/sub-08_T1w_brain.nii.gz
Details: Does not look like a BIDS file: /path/to/mask

This error does not occur with FSLeyes 0.31.2+build0. Can checking
BIDS format be suppressed via an option? I looked, as did several
others here, and we aren't finding this settable.

It doesn't seem like an image viewer/manipulator should really care
what the format of the image filenames is.

This situation surprised many of us here, and I am able to replicate easily.

We can make the error stop by removing all .json files from the root
of the BIDS directory, and effectively de-BIDSifying it.

It also appears to work if the image is removed entirely from the BIDS
directory, but I think that would not be obvious, nor in agreement
with current, standard practice according to FSL tutorials, which I
think suggest creating the mask in the same directory.

Assuming that the directory that contains the subject's anatomical
image is username/ds102/sub-08/anat, then what directory location and
filename will FSLeyes accept according to its BIDS validation rules?

I tried creating inside the derivatives directory a directory for FSL,
adding a dataset_description.json per the description at

https://bids-specification.readthedocs.io/en/derivatives/05-derivatives/01-introduction.html

and putting the mask file into
derivatives/fsl/sub-08/anat/sub-08_T1w_brain.nii.gz and that did not
help.

I see from the Changelog at
https://github.com/pauldmccarthy/fsleyes/blob/master/CHANGELOG.rst
that

Metadata from JSON sidecar files (e.g. in BIDS data sets) is now
displayed in the overlay information panel.

Anyway, if you would, please let us know where FSLeyes expects to see
the brain masks generated by bet/bet2, which JSON files might need to
be written, and what information might be required in them.
Presumably, this will also be a problem when using FSLeyes to look at
images directories created by feat, if they are in the same BIDS
folder?

Thanks in advance for your guidance, -- bennet

########################################################################

To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1


________________________________

To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1

########################################################################

To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1


To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1