Re: [fluka-discuss]: reduction of number of materials while using dicom images

From: Thiago Viana Miranda Lima <>
Date: Tue, 27 Oct 2015 12:31:11 +0000

Hi Minal,

Flair/Fluka is not capable to define regions in the same manner as you are describing. You can, as Wioletta, mentioned use the Dicom tool to correlate the different CT number to a different composition. This will create regions and assignment directly. So you do not need to manually input what is what in the regions. If you open the vxl file this assignment is already present there. Have a look at Schneider's segmentation paper [1]. After the MC run, you would have to do a post-processing to segment the regions you are interested.

One suggestion, if you have for example the delimitation of the different organs (in radiotherapy it is in the RTstruct file) you can sample the positions of each organ on the final usrbin results and display individual organs results.

Best wishes

[1] Schneider W, Bortfeld T, Schlegel W. Correlation between CT numbers and tissue parameters needed for Monte Carlo simulations of clinical dose distributions. Phys. Med. Biol. 2000;45:459–78. [PubMed<>]

On 27 Oct 2015, at 12:45,<> wrote:

Hi Wioletta,

The dicom/information tab gives number of regions as 3035. These are the
regions identified by flair. But I will have to assign the regions to each
of these regions in assignment card. How I will able to do unless I know
which region is having which region number.

I want the details of the regions in each CT slice so that i can have an
array of voxels which are identified with particular organ id. so that i
can use each region for simulating internal source.

I want to know which is bone, muscle or adipose etc and identify each
voxel with its organ id.

I hope now my point is clear.

Minal Y.

Dear Minal,

Each region represents the group of voxels with the same HU value. The
number of unique HU you can check in the DICOM/Information Tab.
While choosing the .mat file, you are actually defining the range of HU
to which the specified material is assigned, thus typically several
voxel regions are sharing the same material. Reducing the materials in
.mat file, limits only the number of possibilities to be assigned, not
the number of regions.

Currently in flair there is no easy way to visualize the voxels
belonging to the same regions. However, you can use for this purpose
Geometry viewer. By selecting "Material" option from the upper menu, you
can change the color for each material for better enhancement and then
investigate the Lattice or 3D visualisation.

Hoping this answer your question.

From:<> [<>] on
behalf of<> [<>]
Sent: 23 October 2015 07:19
Subject: [fluka-discuss]: reduction of number of materials while using
dicom images


I was trying to make voxel phantom using dicom images. I could able to
generate .vxl file. But it makes thousands of regions depending upon the
number of voxels.

By going through the threads i realised that it is possible to only
reduce the number of materials. But how that can be done?

1. I have tried using a modified .mat file containing few materials by
specifying the ranges of HUs. But it is still the same number of

2. There may be large number of voxels belonging to same region, how to
identify them using flair?

Kindly help as the information on this topic is not available.

Thanks in advance.

Minal Y.
Scientific officer,
Internal dosimetry section, BARC

You can manage unsubscription from this mailing list at

You can manage unsubscription from this mailing list at

You can manage unsubscription from this mailing list at
Received on Tue Oct 27 2015 - 14:57:17 CET

This archive was generated by hypermail 2.3.0 : Tue Oct 27 2015 - 14:57:18 CET