[Camino-users] datasynth ERROR on sequence parameters

[Slater David]

Hi Alessandro,

I think your problem is coming from the scheme file you are using. Try creating a Stejskal-Tanner scheme as described here: http://cmic.cs.ucl.ac.uk/camino//index.php?n=Docs.SchemeFiles<http://cmic.cs.ucl.ac.uk/camino/index.php?n=Docs.SchemeFiles>. It should take the following form:

VERSION: STEJSKALTANNER
x_1 y_1 z_1 |G_1| DELTA_1 delta_1 TE_1
x_2 y_2 z_2 |G_2| DELTA_2 delta_2 TE_2
:
:
x_N y_N z_N |G_N| DELTA_N delta_N TE_N

Save it as something like ‘myscheme.scheme’. This type of scheme file contains the gradient pulse information so you can remove these options from the datasynth command. Thus the following command will output synthetic data in .txt format:

datasynth -schemefile myscheme.scheme -voxels 10000 -walkers 100000 -tmax 1000 -geometry cylinder -packing SQUARE -cylinderrad 1.9E-6 -cylindersep 4E-6 -diffusivity 2E-09 -noisetype rician -snr 16.0 -p 0.0 -initial uniform  | float2txt > SIMdata.txt

Note that I have removed the –G, –del and –DEL options which were causing you errors and used the float2txt command to get the output in .txt format.

Try this with a corrected scheme file and you should get your synthetic voxel data.

Best,
David


De : camino-users-bounces at www.nitrc.org [mailto:camino-users-bounces at www.nitrc.org] De la part de Alessandro Calamuneri
Envoyé : 16 October 2014 12:21
À : camino-users at www.nitrc.org
Objet : [Camino-users] datasynth ERROR on sequence parameters

Hi camino experts,

I need help becaues i am not able to create synthetic data using datasynth command; so far I am basically following examples within the command itself.

At this step I am simulating a number of voxels using a cyilinder geometry with square packing. This is what is written in the manual of datasynth

datasynth -schemefile tensorDirs_b=500.scheme -voxels 10000 -walkers 100000 -tmax 1000 -p 0.0 -initial uniform -snr 16.0 -G 0.022 -del 0.032 -DEL 0.04 -geometry cylinder -packing SQUARE -cylinderrad 1.9E-6 -cylindersep 4E-6 > MCdata.Bfloat

which I suppose having something strange, because schemefile was not written out; furthermore there is a tensorDirs_b option that I was not able to understand, it is not on the list. What does it stand for?

However, looking at the example scanner parameters details, I have created a scheme file with 10 b0 acquisitions and 61 diffusion weighted directions and the same bvalue as estimated from parameters inserted.

Thus, I type

datasynth -schemefile scheme.scheme -voxels 10000 -walkers 100000 -tmax 1000 -G 0.022 -del 0.032 -DEL 0.04 -geometry cylinder -packing SQUARE -cylinderrad 1.9E-6 -cylindersep 4E-6 -diffusivity 2E-09 -noisetype rician -snr 16.0 -p 0.0 -initial uniform > MCdata.Bfloat

But I get the following error

Oct 16, 2014 10:49:06 AM tools.CL_Initializer checkParsing
WARNING: WARNING: couldn't parse arg 10: '-G'
Oct 16, 2014 10:49:06 AM tools.CL_Initializer checkParsing
WARNING: WARNING: couldn't parse arg 11: '0.022'
Oct 16, 2014 10:49:06 AM tools.CL_Initializer checkParsing
WARNING: WARNING: couldn't parse arg 12: '-del'
Oct 16, 2014 10:49:06 AM tools.CL_Initializer checkParsing
WARNING: WARNING: couldn't parse arg 13: '0.032'
Oct 16, 2014 10:49:06 AM tools.CL_Initializer checkParsing
WARNING: WARNING: couldn't parse arg 14: '-DEL'
Oct 16, 2014 10:49:06 AM tools.CL_Initializer checkParsing
WARNING: WARNING: couldn't parse arg 15: '0.04'
Exception in thread "main" java.lang.ClassCastException: imaging.B_VectorScheme cannot be cast to imaging.SimulableScheme
                at tools.CL_Initializer.initDataSynthesizerFromSimulation(CL_Initializer.java:2031)
                at tools.CL_Initializer.initDataSynthesizer(CL_Initializer.java:2121)
                at apps.SyntheticData.initOptions(SyntheticData.java:55)
                at apps.Executable.<init>(Executable.java:24)
                at apps.SyntheticData.<init>(SyntheticData.java:34)
                at apps.EntryPoint.getExecutable(EntryPoint.java:222)
                at apps.EntryPoint.main(EntryPoint.java:282)

There is something wrong with the scheme, as I do not see mistyping errors.
I then discarded my scheme file and created a scheme on the fly using options fixedbvalue and fixedqmod, hence the following

datasynth -fixedbvalue 10 61 1.0124E+09 -voxels 10000 -walkers 100000 -tmax 1000 -G 0.022 -del 0.032 -DEL 0.04 -geometry cylinder -packing SQUARE -cylinderrad 1.9E-6 -cylindersep 4E-6 -diffusivity 2E-09 -noisetype rician -snr 16.0 -p 0.0 -initial uniform> MCdata.Bfloat

datasynth -fixedmodq 10 61 1.8578E+05 0.04 -voxels 10000 -walkers 100000 -tmax 1000 -G 0.022 -del 0.032 -DEL 0.04 -geometry cylinder -packing SQUARE -cylinderrad 1.9E-6 -cylindersep 4E-6 -diffusivity 2E-09 -noisetype rician -snr 16.0 -p 0.0 -initial uniform> MCdata.Bfloat

Same error occur. What is wrong?
As far as I know, bvalue is given by the formula

b=(gamma*delta*G)^2*(DELTA-delta/3)=q^2*(DELTA-delta/3)
where gamma should be 42e6*2*pi. All units follow SI

Can you help me understanding why the command is not running? Is my formula wrong?

Best,
Alessandro


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.nitrc.org/pipermail/camino-users/attachments/20141016/71fffa53/attachment-0001.html>