[Gate-users] NCAT Source-phantom

Fri Jan 22 16:43:11 CET 2010

Hello,

Ok.

 My phantom and source are the same size for sure because is the same
interfile and I put the source a half from the phantom in a negative
direction axis, I just define different data on range tables for att or act
table that call for the same image.

My phantom body finishes in the end of the image. Do you think it is
possible to be the problem? I think not because is the same phantom from I
used in Gate4 and it didn’t have problems before.

If I try to use compressed function Do you think it can resolve?

thanks

From: Simon Stute [mailto:gate.stute at gmail.com] 
Sent: sexta-feira, 22 de janeiro de 2010 13:23
To: Cláudia Régio Brambilla
Cc: Ida Häggström; gate-users at lists.opengatecollaboration.org
Subject: Re: [Gate-users] NCAT Source-phantom

Hi again,

2010/1/22 Cláudia Régio Brambilla <claudinha.rb at gmail.com>

Hello,

I was using gate4 when I did that script with parametrizedBoxMatrix. When
gate was released for gate5 I just take the same script change the commands
for regular Matrix and ran. 

OK I understand, I forgot that there was no regularMatrix in Gate4.

The problem starts to appeared after that. In gate4 I’ve never seen this
error message with my simulation.

It is normal since this error message is on related to the RegularNavigator
based on the PhantomParameterization in G4 (only used for this navigator).

We are trying to run in another computer to check if there is a problem with
our installation. Our G4 version is 4.9.2.p01 with gate5.

It is OK, because in the 9.1 version there was a bug in the G4 code in the
GetReplicatNo function in the PhantomParameterization class. But it was
corrected for the 9.2 version.

I think the last one was G4.9.2 and gate 4, I’m not have sure I can check
and reply for you again.

It is not important anymore, because you used the parameterizedBox only with
Gate4.

So to sum up, your problem may come from a wrong definition of your
voxelized volumes (attenuation and/or sources) that did not cause trouble
using the parameterizedBoxMatrix but not with the regularMatrix. Do your
source is has exactly the same dimensions as your phantom ? If you have a
voxelized phantom, your source has to be equal or smaller than this phantom
(analytical or not). Because if it is bigger, some particle may born between
the end boundaries of your phantom box and the distance tolerance in G4
(some nanometer I think). I think in this case you will get this error.
Otherwise, I don't know what is the cause of your problem ! I use regular
and fictitious navigators every time and I only met this error when
developing the regularNavigator in GATE.

Regards,
Simon

Regards

Cláudia

From: Simon Stute [mailto:gate.stute at gmail.com] 
Sent: sexta-feira, 22 de janeiro de 2010 09:49
To: Cláudia Régio Brambilla
Cc: Ida Häggström; gate-users at lists.opengatecollaboration.org

Subject: Re: [Gate-users] NCAT Source-phantom

Hi,

I did not understand that this is a systematic error ! in this case it is
not a rounding error, something goes wrong in your simulation definition I
think.
But it is strange that your macro ran fine in Gate4 and not anymore in
Gate5...

2 questions:

 - If you really change nothing in your macro (I mean really nothing) and
you launch it using Gate4, does it work ?
 - which G4 versions are you currently using with Gate4 and Gate5, the same
or different ? and which ?

Cheers,
Simon

On Thu, Jan 21, 2010 at 8:49 PM, Cláudia Régio Brambilla
<claudinha.rb at gmail.com> wrote:

Hello,

Thanks for clarify my question, I didn’t know about this problem in G4. I’m
using a random seed (auto) and all runs have been  with the same problem.
What Do you suggest for this problem? Is Just to run many times after bug
until work? Or Is the best way to change the regular matrix for compress or
another compatible command?

Thanks all  for all suggestions

Best regards

Cláudia

From: Simon Stute [mailto:gate.stute at gmail.com] 
Sent: quinta-feira, 21 de janeiro de 2010 12:31
To: Ida Häggström
Cc: Cláudia Régio Brambilla; gate-users at lists.opengatecollaboration.org
Subject: Re: [Gate-users] NCAT Source-phantom

Hi all,

This bug comes from the G4PhantomParameterization class of G4 that is used
when using the G4RegularNavigator (it is the case only when you use the
regularMatrix object in GATE, it is not possible for other types of
volumes).
And unfortunatly it is probably a rounding bug when some particle are
generated or interact exactly on the phantom border within the distance
tolerance of G4.
This bug has been already tracked but no satisfying solution was found !
And there is nothing related to the version of Gate in this problem.

Just launch again your simulation with another random seed...

Cheers,
Simon

2010/1/21 Ida Häggström <ida.haggstrom at radfys.umu.se>

Dear Cláudia,

I've seen that error message myself and have not figured out exactly why it
appears. I got it when I used a voxelized phantom based on
"parameterizedBoxMatrix" in Gate 5 (never used Gate 4). When I switched to
"compressedMatrix" instead, everything else exactly the same, the error
disappeared. You can give it a go! =) You can also try "regularMatrix".

/Ida

2010/1/20 Cláudia Régio Brambilla <claudinha.rb at gmail.com>

> Hello Gate Users,
>
>
>
> I was running a simulation with NCAT phantom and this error starts to
> appear
> using GATE 5. I usually used the same code in GATE 4 that any errors
> appeared before.
>
> Does anybody know about the error bellow?
>
>
>
>
>
> ERROR - G4PhantomParameterisation::GetReplicaNo()
>
>        localPoint - (-1.743113937,-1.77535309,-1.503948302) - is outside
> container solid: NCAT_S
>
>
>
> *** G4Exception : InvalidSetup
>
>      issued by : G4PhantomParameterisation::GetReplicaNo()
>
> Point outside voxels!
>
> *** Fatal Error In Argument *** core dump ***
>
>
>
> *** G4Exception: Aborting execution ***
>
> Aborted
>
>
>
>
>
> __________________________________
>
>
>
> Cláudia Régio Brambilla
>
>
>
> Física Médica
>
> Supervisora de Radioproteção - CNEN FM-0089
>
>
>
> Mestranda em Clínica Médica - FAMED/PUCRS
>
> Grupo de Pesquisa em Imagens Médicas - NIMed - PUCRS
>
> TECNOPUC 96A
>
>
>
> Contato:
>
>
>
> +55 51 84251368
>
> +55 51 3320 3682 ramal 7813
>
>
>
> _______________________________________________
> Gate-users mailing list
> Gate-users at lists.opengatecollaboration.org
> http://lists.opengatecollaboration.org/mailman/listinfo/gate-users
>
_______________________________________________
Gate-users mailing list
Gate-users at lists.opengatecollaboration.org
http://lists.opengatecollaboration.org/mailman/listinfo/gate-users