This project has moved. For the latest updates, please go here.

How many photons should I run, that is, what should I set "N" in the infile to?

You want to generate results that have a relative error < 3-5%.  Relative error is the Standard Deviation divided by the Mean.  To generate Standard Deviation estimates, set "TallySecondMoment" to true in the Simulation Options portion of the infile and run the simulation.  In the MATLAB script files that post-process the MCCL results, the Standard Deviation is determined but not plotted.  In the script "load_results_script", you can access the these values by "results{di}.ROfRho.Stdev" if, for example, you specified an ROfRho detector in your infile.

 

How do I specify a semi-infinite layered medium?

Currently MultiLayerTissue assumes an air layer above and below the tissue layers.  However, you can make the bottom tissue layer as thick as you'd like.

 

How do I know how many photons are lost out the bottom of the tissue?

Adding TDiffuseDetectorInput into the list of Detector Inputs in the infile will provide output of the photon weight lost out the bottom of the tissue.

 

Can I specify the same detector twice in the infile?

Yes!  This is useful when you'd like to determine R(rho) for various rho bin sizes, numbers or spans.

 

What is the implementation of detectors as a function of rho?

Detector results as a function of rho, e.g. ROfRho, are determined using concentric circle surface detectors.  In the infile, specifying a Rho Start=0, Stop=10 and Count=101 designates 100 (1 less than Count) surface rings with radii going from [0-0.1]mm, [0.1-0.2]mm, ..., [9.9-10.0]mm. The weight of all photons collected out a ring are summed and normalized by the total number of photons launched and the surface area of the ring.  This provides a reflectance estimate designated at the midpoint of the ring, per square mm.

Note that to use rho-dependent detectors, the source and tissue definitions need to be axi-symmetric about the z-axis.  Therefore, an angled source defined in the infile along with an ROfRho detector input will not make sense.  We have validation checks in the software to make sure the source, tissue and detectors are consistently defined.

Last edited Jan 14, 2013 at 9:25 PM by hayakawa, version 1

Comments

No comments yet.