[Gate-users] Standard HU ranges for Abdomen and Brain CT images

Thu Apr 8 12:15:40 CEST 2021

Hi Atiq,

do you know these papers:

Schneider, W., Bortfeld, T.,  &  Schlegel, W. (2000). Correlation between CT numbers and tissue parameters needed for Monte Carlo simulations of clinical dose distributions. Physics in Medicine and Biology, 45(2), 459–478.

http://www.ncbi.nlm.nih.gov/pubmed/10701515

Kanematsu, N., Inaniwa, T.,  &  Nakao, M. (2016). Modeling of body tissues for Monte Carlo simulation of radiotherapy treatments planned with conventional x-ray CT systems. Physics in Medicine and Biology, 61(13), 5037–5050.

https://doi.org/10.1088/0031-9155/61/13/5037

... those are good ones to read I think.

There are different ways to implement the HU-tissue conversion. A piece-wise constant conversion is one way, i.e., use the same tissue, same density etc for any HU within a certain interval. That yields locally "constant" voxel geometries and is what you are concerned about. You can also fix the tissue composition in a given HU interval and scale the mass density to make the material properties vary effectively within the interval and obtain the desired transitions between HU intervals. Or you create binary mixtures of two tissues by interpolation.

I am not sure how GATE handles that by default because I have not used the CT feature in a while. Others can explain that to you much better.

I think there is no general standard about the conversion procedure from HU to tissue and the intervals to be used. Radiotherapy centers all have there own protocols, all more on last based on literature and best practice.

In case you want to compare your simulation to experimental data, bear in mind the the HU in your image depend on the CT scanner, the reconstruction protocol and potentially on the dimensions of your object (due to beam hardening).

Hope that helps

Nils

On Apr 8 2021, at 11:33 am, Atiq Ur Rahman  <atiqchep at gmail.com>  wrote:

>   
>   
> Hi Maria   Inês,
>   
> Thank you so much for your feedback.   
>   
>   The purpose of using CT is to incorporate   inhomogeneity   in the human body and targets.    In my opinion, setting long intervals may affect the actual motive. otherwise, the material may become a phantom of discrete   materials   (which also become in case of very small bins too) with sharp variations on interfaces of the HU which may take the whole practice   of simulation very   far from the real situations.     
>   
> Secondly, I have a question about the HU unit ranges. By the way, from which resource do you use these range numbers for different organs.   Do you have a standard list for that? Do you mind sharing this information or point to me!    Because my background knowledge about CT and Human organs is really poor!    How can I solve this problem?     
>   
> Regards
>   
> Atiq
>   
>   
>   
>   
>   
> On Thu, Apr 8, 2021 at 4:57 PM Maria Inês Ribeiro  <mid.ribeiro at campus.fct.unl.pt (mailto:mid.ribeiro at campus.fct.unl.pt)>  wrote:
>   
>   
> >   
> >   
> > Hi Atiq,   
> >   
> >
> >   
> > I am doing something similar for dosimetry purposes.   
> >   
> > From what   I have read, the HU depends on the CT scanner that acquired the images. For dosimetry, it is not necessary to produce such small bins (as in your example) because there is not a good relationship   between CT unit and tissue composition in the range of soft tissues and dose does not differ much within them.   
> >   
> >
> >   
> > I would let the experts add more to what I said.   
> >   
> >
> >   
> > Best,   
> >   
> >
> >   
> > Inês
> >   
> >   
> >   
> >   
> >   
> > Atiq Ur Rahman  <atiqchep at gmail.com (mailto:atiqchep at gmail.com)>  escreveu no dia quinta, 8/04/2021 à(s) 08:47:
> >   
> >   
> > >   
> > >   
> > > Dear    experts,
> > >   
> > > I am trying to read the CT images in Gate. I am successful   to read the CT Image but I am not sure about the precise organ density ranges in HU units for Abdomen and Brain CT images. Right now, I am trying to implement it from different online research papers and websites but that may not be a standard Benchmark.
> > >   
> > > For example, I am using this file for Abdomen CT:
> > >   
> > >   
> > > -1024    -600 Air
> > >   
> > > -600     -120     Adipose
> > >   
> > > -120     -90        TissueEquivalent
> > >   
> > > -90         1         Blood
> > >   
> > > 1            10         Mucus
> > >   
> > > 10         20         Intestine
> > >   
> > > 20         30         Kidney
> > >   
> > > 30        40            Pancreas
> > >   
> > > 40         50         Spleen
> > >   
> > > 50         200        Liver
> > >   
> > > 200        400        SpangyBone
> > >   
> > > 400       3072    SpineBone
> > >   
> > >   
> > >   
> > >   Can anyone point me to some standard material   conversion files or resources with HU ranges?       
> > >   
> > >   
> > > Looking forward to any feedback.
> > >   
> > > Regards
> > >   
> > > Atiq   
> > >   
> > >   
> > > _______________________________________________
> > >   
> > > 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
> > >   
> >   
> >   
>   
>   
>   
> _______________________________________________
>   
> 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/pipermail/gate-users/attachments/20210408/a2aaee22/attachment.html>