Diffusion MRI simulation in JEMRIS

[Md. Raiyan Chowdhury]

Dear Tony et al. ,

At first, accept my gratitude for developing JEMRIS as a tool for MRI simulation. It is an excellent tool.

I have been trying to simulate Diffusion MRI, I have tried to find documentation on this topic but could find very few. To be specific, I only find some explanation in a lecture at www.esmrmb.org

I am still not sure about the underlying concept of diffusion MRI simulation in JEMRIS. I have designed some simple phantoms which are 2D static images. My question is, can I use a dynamic phantom in JEMRIS.? By 'dynamic' I meant an animated phantom, to my opinion I think that an animated phantom can resemble diffusion more accurately.

Please let me know your opinion. Thank you.

[Daniel Pflugfelder]

Hi Raiyan,

 

I was still working on Diffusion simulations when I left the Institute about 7 years ago. There are some functionalities implemented, but this part of JEMRIS is certainly not as complete as one might wish for. As far as I know this part has since then not been developed further.

The ESMRMB lecture you already found is all the documentation that exists. In its current form, diffusion simulation in JEMRIS aims at simulating the MRI signal, not imaging sequences. Basically the idea is to set up a microstructure in which the spins can diffuse. This box can be empty for free diffusion, or some basic shapes like cylinders or boxes can be included to simulate restricted diffusion. This allows some simulations, but for more advanced things such as e.g. simulating a DTI sequence with different tensors in each voxel you would need to further develop the source code of jemris. In the past e.g. Alexandre Fortin did this for angiography simulations.

 

Concerning the 'animated phantom': The phantom can be moved during the simulation by supplying a motion trajectory. This will result in a rigid motion of the sample. For details you can also consult the mentioned ESMRMB lecture.

 

As you asked for my opinion on this: You can do some diffusion signal simulations, but if you want to simulate diffusion imaging sequences with JEMRIS this will not run out of the box. You need to be willing to invest time, manpower and you will need programming skills. As far as I know the mentioned angiography project was a PhD thesis.

 

Cheers,

Daniel

 

 

Gesendet: Mittwoch, 09. Dezember 2020 um 23:48 Uhr
Von: "Md. Raiyan Chowdhury" <raz...@gmail.com>
An: "JEMRIS" <jem...@googlegroups.com>
Betreff: Diffusion MRI simulation in JEMRIS

--
You received this message because you are subscribed to the Google Groups "JEMRIS" group.
To unsubscribe from this group and stop receiving emails from it, send an email to jemris+un...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/jemris/ebdeac5d-5076-42c2-a2b9-760a2f9d7d8bn%40googlegroups.com.