the attached images show two explames of the kinds of failrures I see, the red overlaps ay the c1 files (the grey matter). 

Browsing old posts in the list gives the suggestion that non-uniformity in the data may be the source of the problem. I thought maybe there was some hyper-signal in the back of the skull, but I tried running BET first (skull stripping), which made no difference. I also tried correcting for inhomogeneities in the data with no success. 

Any suggests on possible causes and solutions?

Thanks

Colin