Dear experts,
I question concerning neg eigenvalues from DTIFit
I am interested in looking into all the DT eigenvalues. However, when
running DTIFit I get some occasion of neg eigenvalues in lambda_3 (L3)
in all my data sets. These neg eigenvalues are only see in the in the
genu and splenium of the CC.
Read in an old post that DTIFit would be upgraded from just log-linear
solver to account for possible erroneous neg eigenvalues.
My question is, has this been done? Or do I have bad data in these
voxels? Or is this just an effect of the tract characteristics in the
bundle and that the solver have problems with?
Kind regards,
Finn L
|