[Gate-users] Tips&Tricks regarding Voxelized Sources

Fri Mar 20 17:22:17 CET 2015

Dear David

I can confirm, there are still issues in the use of interfile
voxelized sources, however there are no issues at all if you use
mhd/raw images for voxelized sources.

Regards
Alex

2015-03-20 9:47 GMT-04:00, David Sarrut <David.Sarrut at creatis.insa-lyon.fr>:
> Thanks for sharing !
>
> I do not use interfile myself, but one information could be useful: in the
> new 7.1 version (in the develop branch), voxelized source can now be
> .mhd/raw images which could simplify some issues encountered with
> interfile. This is a first version and particularly difficult to test, so I
> would be happy if someone could report (hopefully) successful or
> (unfortunately) unsuccessful use.
>
> Note that the coordinate system with voxelized source is not consistent
> with all other images coordinate systems: it is not centred in the image,
> but in the corner. This may change in the future.
>
> Sincerely,
> David
>
> On Fri, Mar 20, 2015 at 9:55 AM, J.J.M. Kouwenberg <
> j.kouwenberg at erasmusmc.nl> wrote:
>
>> Dear Gergely,
>>
>> This was done on the latest vGate, so Gate v7.0. It might still very well
>> be that I made an error somewhere. These settings just made it work for
>> me
>> on both Gate 7.0 and Gate 6.2.
>>
>> With kind regards,
>>
>> Jasper J.M. Kouwenberg
>> Medical Physics Graduate student
>> Nuclear Medicine
>>
>> E-mail j.kouwenberg at erasmusmc.nl | Phone +316 2532 9809
>>
>> ________________________________________
>> From: gate-users-bounces at lists.opengatecollaboration.org [
>> gate-users-bounces at lists.opengatecollaboration.org] on behalf of Patay
>> Gergely [patay.gergely at mediso.hu]
>> Sent: 20 March 2015 09:49
>> To: Gate User
>> Subject: Re: [Gate-users] Tips&Tricks regarding Voxelized Sources
>>
>> Dear Jasper,
>>
>> concerning the Interfile-related issues: did you use v7.0, or code from
>> the
>> development git branch? Since the 7.0 release there were some
>> enhancements
>> in
>> the Interfile reader code (e.g. correct filenames, endianness and
>> byte/value),
>> and I wonder whether it is still buggy or not.
>>
>> best regards,
>>   Gergely
>>
>> --
>> Gergely Patay
>> physicist
>> Mediso Medical Imaging Systems Ltd.
>> Hungary
>>
>>
>>
>> On 2015-03-19 17:54, J.J.M. Kouwenberg wrote:
>> > Dear all,
>> >
>> > After many hours of frustrated keyboard smashing, I managed to find a
>> number of solutions to undocumented difficulties regarding voxelized
>> sources by delving through the GATE source code.
>> > In order to save some of you a lot of time, I made a small list with
>> tips&tricks below. I'll consider contributing to the Wiki when I get
>> positive responses to this mail.
>> >
>> > Voxelized sources in GATE:
>> >   - It seems that voxelized sources only work when combined with
>> voxelized phantoms. The names of the raw data files are determined adding
>> respectively _atn_[frame number].bin and _act_[frame number].bin to the
>> parameter '/gate/RTVPhantom/setBaseFileName [name]'. The datafile name is
>> the header file does not influence this!
>> >   - The Interfile format for GATE is somewhat peculiar. Instead of
>> 'matrix size [3]' and 'scaling factor (mm/pixel) [3]' for the z-axis, it
>> requires respectively 'number of slices' and 'slice thickness (pixels)',
>> where the latter is still given mm.
>> >   - While the Interfile offers an 'imagedata byte order' option, it
>> > does
>> not seem to affect the actual readout. The default is BIGENDIAN with 16
>> bits per value (2 bytes). One can alter this in
>> GateSourceVoxelInterfileReader.cc. Note that the _RT_ functions are used
>> instead of the regular functions!! To make sure your file is read
>> correctly, check the raw data file with a binary reader (GHex e.g.) and
>> compare it to the 'imageValue' values in
>> GateSourceVoxelInterfileReader.cc.
>> >   - When the voxel reader returns zero voxels (number of voxels: 0), it
>> means that none of the voxel values where converted to an activity higher
>> than zero becquerel. It does not mean that you data file is faulty, but
>> more likely that there is an error in the translation.
>> >   - The maximum number of subdivisions in your TimeActivity table is
>> > 400
>> (at least on the vGate system). Higher values give a segment break error.
>> >   - The 'SetTimeSampling' is not the same as the spacing in your
>> TimeActivity table. The code interpolates the table using the
>> SetTimeSampling interval and a B-Spline. This can be changed in
>> GateVSourceVoxelReader.cc.
>> >   - The code for the voxelized source is quite talkative. Many of the
>> outputs are coded in the two mentioned source files and are unaffected by
>> verbosity. It is highly recommended to comment out the activity curve
>> filling output when using large tables.
>> >   - When using STIR's generate_image to generate a source, the
>> > following
>> Interfile settings are recommended:
>> > "output file format type:= Interfile
>> >    interfile Output File Format Parameters:=
>> >      number format := unsigned integer
>> >      number_of_bytes_per_pixel:=2
>> >      scale_to_write_data:= 1
>> >      byte order:=BIGENDIAN
>> >    End Interfile Output File Format Parameters:="
>> >
>> > This is all for now.
>> >
>> > With kind regards,
>> >
>> > Jasper J.M. Kouwenberg
>> > Medical Physics Graduate student
>> > Nuclear Medicine
>> >
>> > E-mail j.kouwenberg at erasmusmc.nl | Phone +316 2532 9809
>> > _______________________________________________
>> > 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
>> _______________________________________________
>> Gate-users mailing list
>> Gate-users at lists.opengatecollaboration.org
>> http://lists.opengatecollaboration.org/mailman/listinfo/gate-users
>>
>
>
>
> --
> David Sarrut, Phd
> Directeur de recherche CNRS
> CREATIS, UMR CNRS 5220, Inserm U 1044
> 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
> _________________________________
>  "2 + 2 = 5,  for extremely large values of 2"
> _________________________________
>