I've just modified the way the DEV version of mri_convert reads in generic (ie, non-siemens) dicom files. This change makes it so that it can convert multi-frame GE (and possibly other) dicom files. Thanks to Adam Thomas at NIH for supplying the GE data.
I've also fixed a slight bug in the way the coordinates in the non-siemens dicoms were read (they were off by 0.5 voxel). I suspect that no one will notice this change unless you do a "diff" between the same dicom converted with the different versions.
If you want to use the old dicom reader, run mri_convert with -dicomread0.
Please let me know if you have any questions or run into any problems. Also let me know if you have some data you'd like me to test it out on. I have not tested it out on diffusion data yet, for exmaple.
doug