[Gate-users] Regular Parameterization Error with Geant4 v10

Mon Feb 13 17:07:55 CET 2017

Hello Nicolas,

Yes I remember this. It is a bug I think.
I just set min, max: -1,4 in my data/waterbottle_HU2mat.txt
And then it doesn’t complain.

This is just a fix and hopefully this bug will be corrected soon.

Cheers,
  Maxime

> On 13 Feb 2017, at 16:38, Triltsch, Nicolas <nicolas.triltsch at tum.de> wrote:
> 
> Hey David & Maxime,
> 
> Thanks for your advice. I tried to use the option ImageRegionalizedVolume (plus the distance map) instead of ImageNestedParametrisedVolume, but it gives the following error:
> 
> "[Physic-0] The following Geant4's physic-list is enabled :emlivermore
> GateVImageVolume.cc (l.363): The image contains HU indices out of range of the HU range found in data/waterbottle_HU2mat.txt
> min, max:0, 4.
> min, max in image: -1, 3
> Abort.
> "
> 
> I am confused by this error as I am using the the same HU2mat.txt and .mhd+.raw files as before. Does the new Parameterization method require different material descriptors or what could be the reason? My image contains of 4 materials only: Air (index 0), water (index 1), plastic (index 2) and lead (index 3).
> Cheers,
> 
> Nico
> On 02/10/2017 04:57 PM, David Sarrut wrote:
>> Hello, 
>> 
>> To my knowledge, RegularParameterization with flag skipEqualMaterials is still buggy (it should be disabled: if someone know this code, please propose a pull request to correct or disable this option). 
>> 
>> Use either: 
>> - ImageNestedParametrisedVolume : simple conventional navigator, similar to regular, (slightly faster)
>> 
>> - ImageRegionalizedVolume as proposed by Maxime. It could be 3-4 times faster in some cases but it depends on what you are storing during the simulation. If it is a dose map, take care to the statistical uncertainties. If you are only interested in particle outgoing the volume (for imaging for example), it is safe to use. Of course, the speed up depends on the voxelized volume (faster if lot of similar material). 
>> 
>> Doc is here: http://wiki.opengatecollaboration.org/index.php/Users_Guide_V8.0:Voxelized_Source_and_Phantom#Nested_parameterization_method <http://wiki.opengatecollaboration.org/index.php/Users_Guide_V8.0:Voxelized_Source_and_Phantom#Nested_parameterization_method>
>> 
>> David
>> 
>> 
>> On Fri, Feb 10, 2017 at 4:45 PM, Maxime Chauvin <maxime.chauvin at inserm.fr <mailto:maxime.chauvin at inserm.fr>> wrote:
>> Hello Nicolas,
>> 
>> I had the same problem as you and couldn’t fix it. On the other hand there is a even more efficient way  using
>> ImageRegionalizedVolume. It will further speed up the simulation especially if your geometry has a lot of common voxel
>> and you will get rid of this warning.
>> 
>> Cheers,
>>   Maxime 
>> 
>>> On 10 Feb 2017, at 14:41, Triltsch, Nicolas <nicolas.triltsch at tum.de <mailto:nicolas.triltsch at tum.de>> wrote:
>>> 
>>> Hello,
>>> 
>>> I tried to speed up my simulation by choosing Regular Parameterization (with flag skipEqualMaterials to 1) instead of Nested Parameterization, because my Volume got a lot of voxels with the same material. It did indeed speed up the simulation by a factor of 15, but some projections of my CT simulation failed (error: segmentation fault; crash).
>>> 
>>> Now I saw that during the initialization phase of gate it states the following:  
>>> ### WARNING ### setSkipEqualMaterials at false !! The Geant4 method is not safe since the release 9.5 - Need to be fixed
>>> 
>>> I am pretty sure that this warning is somehow connected to the warning. Does anyone know if there had been a bug fix for Geant4 v10.2.2?
>>> 
>>> In the appendix there is a screenshot of the error.
>>> 
>>> Cheers,
>>> Nico
>>> -- 
>>> B.Sc. Nicolas Triltsch
>>> Masterand
>>> 
>>> Technische Universität München
>>> Physik-Department
>>> Lehrstuhl für Biomedizinische Physik E17
>>> 
>>> James-Franck-Straße 1
>>> 85748 Garching b. München
>>> 
>>> Tel: +49 89 289 12591 <tel:+49%2089%2028912591>
>>> 
>>> nicolas.triltsch at tum.de <mailto:nicolas.triltsch at tum.de>
>>> www.e17.ph.tum.de <http://www.e17.ph.tum.de/><error1.png><error2.png>_______________________________________________
>>> Gate-users mailing list
>>> Gate-users at lists.opengatecollaboration.org <mailto:Gate-users at lists.opengatecollaboration.org>
>>> http://lists.opengatecollaboration.org/mailman/listinfo/gate-users <http://lists.opengatecollaboration.org/mailman/listinfo/gate-users>
>> 
>> _______________________________________________
>> Gate-users mailing list
>> Gate-users at lists.opengatecollaboration.org <mailto:Gate-users at lists.opengatecollaboration.org>
>> http://lists.opengatecollaboration.org/mailman/listinfo/gate-users <http://lists.opengatecollaboration.org/mailman/listinfo/gate-users>
>> 
>> 
>> 
>> -- 
>> David Sarrut, Phd
>> Directeur de recherche CNRS
>> CREATIS, UMR CNRS 5220, Inserm U1206
>> Centre de lutte contre le cancer Léon Bérard
>> 28 rue Laënnec, 69373 Lyon cedex 08
>> Tel : 04 78 78 51 51 / 06 74 72 05 42
>> http://www.creatis.insa-lyon.fr/~dsarrut <http://www.creatis.insa-lyon.fr/%7Edsarrut>
>> _________________________________
>>  "2 + 2 = 5,  for extremely large values of 2"
>> _________________________________
> 
> -- 
> B.Sc. Nicolas Triltsch
> Masterand
> 
> Technische Universität München
> Physik-Department
> Lehrstuhl für Biomedizinische Physik E17
> 
> James-Franck-Straße 1
> 85748 Garching b. München
> 
> Tel: +49 89 289 12591
> 
> nicolas.triltsch at tum.de <mailto:nicolas.triltsch at tum.de>
> www.e17.ph.tum.de <http://www.e17.ph.tum.de/><waterbottle_HU2mat.txt>_______________________________________________
> Gate-users mailing list
> Gate-users at lists.opengatecollaboration.org
> http://lists.opengatecollaboration.org/mailman/listinfo/gate-users

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opengatecollaboration.org/mailman/private/gate-users/attachments/20170213/4cb557cd/attachment.html>