snip>   
   >   
   > Wiki says that the volume fraction is about 45%, and an erythrocyte is   
   >   
   > about 6 or 8 microns across, so the mean free path is something like   
   >   
   >   
   >   
   > MFP ~ 6 um/(0.45)**1/ ~ 7 um,   
   >   
   >   
   >   
   > which is somewhat unrealistically generous since it ignores the volume   
   >   
   > taken up by the blood cells themselves.   
   >   
   >   
   >   
   > So your 1 mm of blood will scatter the light not once but over 100   
   >   
   > times.  In other words, it's completely diffuse.   
   >   
   >   
   I have had some time to at least present the problem in a more concise way.   
   Let's consider the collection of photons that do through this turbid media, say   
   it's blood. And say we have absorbing particles at 1% volume fraction in there,   
   that we want to measure their number +/- 10%. (no imaging, just this number.)   
      
      
   A diode laser beam goes through 1 mm of this media. Some of the photons scatter   
   out of the optical path (which is the possible paths that can end up at the   
   detector.) These are lost. Some are absorbed by the media.   
      
      
   So we are concerned with the cohort of photons that end up at the PD detector.   
   These photons are all signal photons, because they have had to chance being   
   absorbed by the 1% vol fraction absorbers, and therefore carry signal. So if   
   you look at two similar cuvettes, one with and one without the absorbers, there   
   may be a detectable difference, assuming your fundamental photon budget and   
   noise budgets are not violated. So the question is, is this picture plausible   
   enough to have some expectation of a usable signal?   
      
      
   I believe it will center around the issue of whether there are sources of noise   
    unaccounted for in the above picture. (or lack of signal) Perhaps a scattering   
   media can diffuse around absorbing particles and somehow avoid them, though I   
   don't think so. I did a simple geometric calculatuon of the chance a photon   
   will hit an absorber on its 1 mm journey, assuming no scattering, and it's   
    somewhere 30< > 80 %.   
      
      
   Any comments?   
      
   jb   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)