Hi Rohan
It should not make a huge difference in this case, but I still expect model2 to be better. It can better capture partial volume due to the range of diffusivities it represents, so the b300 should help in this aspect.
Cheers
Stam
On 10 Jul 2014, at 21:20, Rohan Katipally <[log in to unmask]> wrote:
> I am doing probabilistic tractography with diffusion data acquired with a combination of b values at 300 s/mm^2 and 1100 s/mm^2. Should I use the standard monoexponential model when I run BedpostX or the multiexponential model (model 2) since I have multiple b values?
>
> Reading some of "Model-Based Analysis of Multishell Diffusion MR Data for Tractography: How to Get Over Fitting Problems" (Jbabdi et al), it seems like the multiexponential model can avoid overfitting when acquiring data with multiple b-values. However, it is also mentioned that non-monoexponential decay in the diffusion decay curve really only appears at b = 1500 and higher. Since my b-vals are below that, which model is more appropriate for me?
|