This project has moved and is read-only. For the latest updates, please go here.

can Inversely calculate optical properties with measured diffuse reflectance spectroscopy?

Jan 24, 2013 at 2:13 PM

Hi everyone,

Can this sofware calculate the optical properties with measured DRS? and how can i do that?


Thank you very much


Feb 28, 2013 at 10:59 PM
Thank you very much for your interest in our software!
I'm sorry it took so long to respond. I wasn't alerted of your post and just happened to see it later.
We currently do not have the capability to invert R(lambda) measurements to determine optical properties.
It is a capability we hope to add in the future.
Jul 1, 2013 at 3:19 AM
Let me just add to this: we do have a Matlab wrapper package here with a number of forward solver capabilities. I believe the samples include forward models with DRS in homogeneous media. You could use the built-in Matlab 'fminsearch' (or lsqcurvefit, lsqnonneg, etc) method to solve an inverse problem based on the forward models. All that's needed would be to define a function to minimize, which could be a chi-squared (sum of the square of the difference between measurement and forward model prediction).

Jul 4, 2013 at 7:29 PM
I have added two new examples in the matlab interoperability code (vts_solver_demo.m) to: 1) generate and plot R(lambda) at a given rho, and chromophore concentrations assuming a power law for scattering, and 2) solve inverse problem for R(lambda) at given rho. This example solves for concentrations of HbO2, Hb and H20 given simulated measurements created using Nurbs scaled Monte Carlo and inverted using a standard diffusion solution with a point source. I created a script sda_F.m that is the forward solver solution in a format compatible with lsqcurvefit.

I did not push this code to codeplex yet because we just recent pushed, however, I can if there is immediate interest.
Jul 5, 2013 at 4:24 AM
Thanks a lot for your reply and congratulations to you for extending one important capabilities or your MC. Recently, inverse Monce Carlo Method have been the most reliable method to detemine optical properites for in vivo measurement. I am very intested in it and i believe there would be some others are interested in. So I hope you can push this code to codeplex or send me a copy alone. haha.

Jul 5, 2013 at 4:07 PM
Thank you for your continued interest in our software! I will push a new Matlab interoperability download to codeplex sometime next week then. I'll post here again when it is ready for download.
Jul 12, 2013 at 6:37 PM
I just pushed a new download to codeplex. It is called VTS Matlab Version 1.0.13. Hopefully you can expand the example for your particular interest.
Jul 27, 2013 at 1:56 PM
Hi Dr.Hayakawa,
I have tried to use your software to doing the inversely calculation. It was fantastic. Thank you verry much. I want to develop it. I believe i will look for your help hahaha. Thank you again.
Best regards
Zhihai Qiu
Fujian normal univeristy

------------------ Original ------------------