search

daducci / AMICO

Accelerated Microstructure Imaging via Convex Optimization (AMICO) from diffusion MRI data
Other
104 stars 62 forks source link

Question about user defined parallel diffusivity parameter in AMICO NODDI model #114

Closed CallowBrainProject closed 2 years ago

CallowBrainProject commented 3 years ago

Hello, several recent studies have suggested that the standard parallel diffusivity parameter in NODDI for intraneurite diffusion (1.7) is suboptimal for measuring gray and even white matter in some samples (https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0217118 and https://www.nature.com/articles/s41598-019-48671-7).

Is there a way to alter the diffusivity parameters of our model in AMICO for the NODDI model? Thank you for any suggestions you can provide.

Daniel

MarioOcampo commented 3 years ago

Hello @CallowBrainProject , Yes, as it was commented in #97, you can modify the parameters to suit your needs. The diffusivity parameters are dPar and dIso and can be modified with the function set(). Let us know if this does what you wanted.

Best. Mario

ejcanalesr commented 3 years ago

Hi guys, I have an idea but it must be tested. For example, the fitting can be done by using many different parallel diffusivities (one by one). The optimal diffusivity should be the one producing a smaller residual (a better fitting to the data). This way, a different diffusivity per voxel could be used. However, this is going to increase the computation time linearly with the number of diffusivities. Anyway, this is something we may explore in the future, as this could solve the more important theoretical limitation of NODDI. What do you think? All the best, Erick

CallowBrainProject commented 3 years ago

Great,

Thank you very much for letting me know. That is indeed what I was looking for.

Daniel

On Mon, Jun 7, 2021 at 2:58 AM Mario Ocampo @.***> wrote:

Hello @CallowBrainProject https://github.com/CallowBrainProject , Yes, as it was commented in #97 https://github.com/daducci/AMICO/issues/97, you can modify the parameters to suit your needs. The diffusivity parameters are dPar and dIso and can be modified with the function set(). Let us know if this does what you wanted.

Best. Mario

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/daducci/AMICO/issues/114#issuecomment-855647277, or unsubscribe https://github.com/notifications/unsubscribe-auth/AIBERDHIJIF26522YPKBGCTTRRUXZANCNFSM46E36ACA .

-- Daniel Callow Graduate Student, Neuroscience and Cognitive Science COMBINE NRT Fellow Exercise for Brain Health Lab University of Maryland, College Park @. @.>* 443-254-6298