Hi there,

I am wondering if there is any way to get slice timing information for diffusion weighted imaging data acquired on a GE scanner (Signa HDxt). I have attempted to use dcm2niix.

I have followed other threads which have led me to the following pages:
https://github.com/rordenlab/dcm2niix/is...
https://github.com/rordenlab/dcm2niix/tr... 

However my understanding is that whilst the private data block approach can work for GE fMRI data, it is not able to extract slice timing information from GE diffusion datasets. Is there another way forward?

Any help would be much appreciated!

Thanks,
Emily

Hi Emily,
I haven't tried to process GE data in a very long time, but am familiar with how this is handled by Philips. I'm curious what kind of information you are looking for - what can you do with slice timing information for diffusion? Is this for dynamic diffusion measurements (functional) or for motion correction?
There are a couple of things that could throw off the slice timing for diffusion - multiple averages are common, particularly for the images with higher b-values, so the resulting image might be an average from multiple time points. Also, for Philips it's possible to split up the slices in a volume into chunks to keep the TR from being too long, so these concatenations (Siemens) or packages (Philips) would also make the slice timing unpredictable to the casual observer.
One question - what are you using to read the DICOM header?
Regards,
Alex

Hi Alex,

Thank you for your reply, and apologies for my delayed response.

I am hoping to use the slice timing information in order to perform slice-volume motion correction as an option in FSL's Eddy. We have a long TR and so ideally we would be able to correct for motion within each volume. Unfortunately, from what I have read on various posts, the slice timing infomation isn't available for GE diffusion scans and some attempted work-arounds only work with fMRI data (below are some relevant posts for you or any interested reader). 

• https://github.com/rordenlab/dcm2niix/tree/master/GE#slice-timing 
• https://github.com/rordenlab/dcm2niix/issues/311

Thanks for the information re/ how Philips handles this - this is useful to know. I have some newly acquired test data from a Siemens Prisma and the .json file does contain slice timing information which is exactly what I was after - if only GE gave this information!

To convert the data I am using the dcm2niix as part of MRIcroGL. 

Thank you,
Emily