Hey Ash,
thanks for your answer, I had one sign
wrong, that's right, (not a native English/Latin speaker, got confused
over anteror and superior).
I set it up this way, because I wanted
to have the simulation input formatted from an h5 dataset.
So if you have a segmentation and multiple
motion models matching this segmentation, then one can store the geometry
information and the motion fields together in one file, and read it in.
I don't like reading too many template
images in, because I have the feeling that too many files are necessary
which all must match each other to get the simulation running, so to me
it feels uncomfortable. I want one file with everyhing in it.
Best
Johannes
--
Johannes Mayer
Quantitative MRI
Physikalisch-Technische Bundesanstalt
Abbestr. 2-12
10587 Berlin, Germany
[log in to unmask]
phone: +49
30 3481 7233
Please visit www.ptb.de
Von:
"Ashley Gillman"
<[log in to unmask]>
An:
[log in to unmask]
Datum:
06.02.2019 02:53
Betreff:
Re: Example
setup GeometricalInfo
Gesendet von:
"CCP-PETMR
Developers list" <[log in to unmask]>
Hey Johannes,
Sorry for the delay.
Your image looks good. You're right, the
origin is somewhat arbitrary, even in a real acquisition - but especially
so for a virtual phantom. Your choice of the center of the image is a good
one I'd say.
To use this, you would use something like:
size = (nx, ny, nz)
direction = <matrix where each column
is unit vector in direction of axis>
offset = <origin as you specified, the
desired position of the center of voxel (0, 0, 0)>
spacing = <desired voxel size>
And pass these to the constructor.
But I wonder maybe why you need to do this?
Would you be better off loading e.g., a NIfTI file as a template - and
therefore using Richard's code to do the initialisation?
Ash
From: CCP-PETMR Developers list
<[log in to unmask]> on behalf of Johannes Mayer <[log in to unmask]>
Sent: Friday, 1 February 2019 21:10
To: [log in to unmask]
Subject: Example setup GeometricalInfo
Hello everyone,
I wanted to ask for confirmation on how to deal with the VoxelisedGeometryInformation.
I drew up a small image, on what I want to put in (superior is going into
the plane).
Now I wanted to know, if I have this image, what the corresponding parameters
are I would have to pass the constructor of VoxelisedGeometryInformation.
(I saw after drawing this thing, that the matrix is computed automatically,
so that would not have to be supplied)
I have another question, which is: can I freely choose the origin of the
LPS system ( as suggested in the sketch I drew up )
So finally: could anyone comment if those are the correct parameters I
have to supply the constructor with for the 3D image in the sketch?
Best
Johannes
--
Johannes Mayer
Quantitative MRI
Physikalisch-Technische
Bundesanstalt
Abbestr. 2-12
10587 Berlin, Germany
[log in to unmask]
phone: +49 30
3481 7233
Please visit www.ptb.de
To unsubscribe from the CCP-PETMR-DEVEL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP-PETMR-DEVEL&A=1
