You may want to take a step back and look over the system you are simulating to understand the underlying photon transport

Optical fluence is J/cm^2; fluence rate is W/cm^2, neither is in the units of photons per area, so you'll need to add an energy per photon factor at whatever wavelength you're using.  

If you are referring to radiant intensity, the units are W/Sr; if irradiance, W/cm^2.

However, irradiance = fluence rate only when a light beam is orthogonal to the surface in question. 

You may want to step back and look at the original MCML paper by Lihong Wang or take a look at a biomedical optics text describing light transport in tissue- these issues come up in the derivation of the diffusion equation and the discretization of analytic solutions for tomographic inversion (among other topics).

You may need to integrate your analytic solution over each voxel to compare to the MC results.

There are a number of good review papers which may help you including Durduran et al, 2010 Reports in Progress in Physics, Arridge et al., Arridge and Schotland, Jacques and Pogue, etc.

regards,

David

Hi David,

Thanks very much for pointing out above discussions.

Compare deviance of both. Smaller deviance is better.  Moreover,  smaller  posterior SD will be better approximation, which often seen with simulations.

a a khan