RE: [fluka-discuss]: DICOM Import Explanation

From: <me_at_marychin.org>
Date: Thu, 26 Mar 2015 10:42:50 -0700 (PDT)

Acknowledging VOXEL### budgeting and the extra degree of freedom for density,
FLAIR's present implementation is not necessarily a bad compromise. In
EGSnrc/BEAMnrc voxel phantoms (.egsphant) are represented by two complete 3D
matrices, one matrix for material type, another matrix for density. I don't
suppose we prefer that.

That's technicality. If we consider the meaning of HU conversion into material
type and density, it does seem more correct, at least to me, not to jumble the
two. The two are sometimes jumbled, when the density is scaled to 'compensate'
for an accounted difference in material composition. This might work for
electromagnetic showers (to which EGSnrc is restricted) to some extend but not
the wealth of particle and reaction types in FLUKA.

:) mary


> On 26 March 2015 at 06:41 "Bartkoski,Dirk Alan" <DABartkoski_at_mdanderson.org>
> wrote:
>
>
> I think I understand. So there is no way to reduce the number of VOXEL###
> regions, only to reduce the number of materials.
>
>
>
>
> ---------------------------------------------
>
> Dirk A. Bartkoski, Ph.D.
>
> Postdoctoral Fellow | Radiation Physics
>
> [MDACC Logo]
>
> Mays Clinic (ACBP1.2972.1)
>
> 1515 Holcombe Blvd., Unit 1202
>
> Houston, TX 77030
>
> (713) 745-8124 Office | (865) 310-7349 Mobile
>
> dabartkoski_at_mdanderson.org <mailto:dabartkoski_at_mdanderson.org>
>
>
>
> From: Vasilis Vlachoudis [mailto:Vasilis.Vlachoudis_at_cern.ch]
> Sent: Thursday, March 26, 2015 2:51 AM
> To: Bartkoski,Dirk Alan; Mary Chin; fluka-discuss_at_fluka.org
> Subject: RE: [fluka-discuss]: DICOM Import Explanation
>
>
>
> Hi
>
> this is correct, flair will assigning a new region for each hounsfield unit,
> while groups of them will share the same materials.
> The reason is that the additional CORRFACTor has to be applied per hounsfield
> unit to take in to account the
> continuous variation of the density.
>
> The material and corrfactor will be embded in the voxel file.
>
> Vasilis
>
>
> ---------------------------------------------
>
> From: owner-fluka-discuss_at_mi.infn.it <mailto:owner-fluka-discuss_at_mi.infn.it>
> [owner-fluka-discuss_at_mi.infn.it] on behalf of Bartkoski,Dirk Alan
> [DABartkoski_at_mdanderson.org]
> Sent: 25 March 2015 18:05
> To: Mary Chin; fluka-discuss_at_fluka.org <mailto:fluka-discuss_at_fluka.org>
> Subject: RE: [fluka-discuss]: DICOM Import Explanation
>
> It does generate a .vxl. Isn't it supposed to generate the same number of
> voxel-regions as I have specified in my .mat file? For example if my .mat file
> has four HU ranges corresponding to four materials then flair should generate
> six regions VOXEL, VOXEL001, VOXEL002, VOXEL003, VOXEL004, VOXEL005, and
> VOXEL006. Where the last four correspond to my four material regions. This is
> not happening. It is generating a separate region for almost every single HU
> value in the DICOM image.
>
>
>
> Furthermore, I am unclear about the material definitions. Are the material
> definitions that I specify in the DICOM tab with .inp embedded in the .vxl
> file or do I also need to define all of materials in the main input file I am
> using as my FLUKA run input?
>
>
> ---------------------------------------------
>
> Dirk A. Bartkoski, Ph.D.
>
> Postdoctoral Fellow | Radiation Physics
>
> [MDACC Logo]
>
> Mays Clinic (ACBP1.2972.1)
>
> 1515 Holcombe Blvd., Unit 1202
>
> Houston, TX 77030
>
> (713) 745-8124 Office | (865) 310-7349 Mobile
>
> dabartkoski_at_mdanderson.org <mailto:dabartkoski_at_mdanderson.org>
>
>
>
> From: me_at_marychin.org <mailto:me_at_marychin.org> [mailto:me_at_marychin.org]
> Sent: Tuesday, March 24, 2015 5:05 PM
> To: Bartkoski,Dirk Alan; fluka-discuss_at_fluka.org
> <mailto:fluka-discuss_at_fluka.org>
> Subject: Re: [fluka-discuss]: DICOM Import Explanation
>
>
>
> Hi Dirk,
>
>
>
> For voxel geometry we have the following files:
>
>
>
> .inp -- as in all other FLUKA simulations, with or without voxels.
>
>
>
> .mat -- a lookup table so that FLAIR may map your DICOM files to a .vxl file.
> This table is what the medical physics community refers to as the 'conversion
> ramp'.
>
>
>
> .vxl -- generated automatically by FLAIR for us, although in principle one
> can still generate this manually without FLAIR's help, the painful way, in
> which case .mat would not be necessary.
>
>
>
> The material.inp which comes with the standard FLAIR distribution contains
> the MATERIAL and COMPOUND cards only. Therefore, it is only an excerpt. It is
> not yet the complete .inp ready for a FLUKA run. We can see that each SDUM in
> the MATERIAL card corresponds to an entry in the second column of .mat.
>
>
>
> Did you get a newly created file with, .vxl extension, in your directory?
> Organ and voxel-regions numbers are embedded in .vxl. If we let FLAIR does the
> conversion for us, these are nothing we need to worry about. We just need to
> specify the name of the .vxl file as SDUM for the VOXELS card in .inp.
>
>
>
> :) mary
>
>
>
> > >
> > On 24 March 2015 at 09:57 "Bartkoski,Dirk Alan"
> > <DABartkoski_at_mdanderson.org <mailto:DABartkoski_at_mdanderson.org> > wrote:
> >
> > I have been trying to import CT DICOM images using flair. I can import
> > them and look at the slices. I have opened the head.mat and material.inp
> > example files. There are a few things I do not understand about the process
> > as explained in the online "Voxel and Medical Applications" slides.
> >
> >
> >
> > 1. I do not understand the meaning of "voxel-region number". In the .mat
> > file you specify the HU value ranges that belong to each material that is
> > specified in the material.inp file. I do not where or how the voxel-region
> > number is supposed to be assigned.
> >
> >
> >
> > 2. flair is supposed to generate regions based on the organ number. This
> > does not happen. I assume it because I can't figure out how to assign organ
> > numbers.
> >
> >
> >
> > Any clarity or direction to additional tutorial resources or examples
> > would be most appreciated.
> >
> >
> >
> >
> > ---------------------------------------------
> >
> > Dirk A. Bartkoski, Ph.D.
> >
> > Postdoctoral Fellow | Radiation Physics
> >
> > [MDACC Logo]
> >
> > Mays Clinic (ACBP1.2972.1)
> >
> > 1515 Holcombe Blvd., Unit 1202
> >
> > Houston, TX 77030
> >
> > (713) 745-8124 Office | (865) 310-7349 Mobile
> >
> > dabartkoski_at_mdanderson.org <mailto:dabartkoski_at_mdanderson.org>
> >
> >
> >
> > >
>
>
>



__________________________________________________________________________
You can manage unsubscription from this mailing list at https://www.fluka.org/fluka.php?id=acc_info
Received on Thu Mar 26 2015 - 20:08:22 CET

This archive was generated by hypermail 2.3.0 : Thu Mar 26 2015 - 20:08:29 CET