I do not know of any issues specific to GE images. In general, they store very little information in their DICOM files, so they are easy to convert. No conversion tool will be able to determine reliably the slice order and phase encoding direction. This is not a deficiency of the conversion, rather this information is not encoding in the DICOM data
  https://github.com/rordenlab/dcm2niix/issues/163

If you do spot any issues, please report them on github.

As with every user, if you plan to do diffusion tensor imaging processing, I recommend you follow the details in the 'dedicated document' linked at
  https://www.nitrc.org/plugins/mwiki/index.php/dcm2nii:MainPage#Diffusion_Tensor_Imaging
this will make sure the diffusion vector gradients are reported in a way that makes sense to your pipeline. I have made every effort to ensure my software generates FSL-compatible bvecs, but I do not have access to GE or Philips equipment, nor have I examined all the possible permutations for image acquisiton.