On Friday, April 18, 2014 10:34:04 AM UTC-4, gghe...@gmail.com wrote:   
   > On Thursday, April 17, 2014 11:19:21 PM UTC-4, haitic...@gmail.com wrote:   
   >   
   > > On Tuesday, April 15, 2014 2:52:09 PM UTC-4, Phil Hobbs wrote:   
   >   
   > >   
   >   
   >    
   >   
   > >   
   >   
   > > Well, I'm thinking that say 20 discrete points across the NIR band might   
   >   
   > > be enough for discriminant tasks. The idea there is use a stepper,   
   SNIP   
   >   
   >   
   > (I must admit I'm totally confused about what you are trying to do.)   
   >   
   >   
   >   
   > George H.   
      
   Hi George   
   I'm a pattern recognition engineer, mainly, and we are usually trying to find   
   data input which is easy to get. I am interested in weak signals from the pov   
   of biomedical diagnostics and a QM effort I can't go into here.   
   As Phil well knows, the NIR is the "opportunist band" for cheap data on the   
   human body. The visible has too much absorption, and the far IR hard to work   
   with.   
      
   SO, NIR. And FTIR seems to produce better data than dispersive spectography.   
   The peaks are much sharper and better defined. But Phil is leery of mechanical   
   movements, fair enough.   
   I'm agnostic.   
      
      
   In my mind, the question of interest is the following: In a welter of signals   
   coming from NIR reflectometry, for instance, is there a signal processing   
   method, after the fact, that will pull out signals buried in the "noise." ?   
      
      
   Yes, you can do things on the front end - polarimetry to remove skin   
   reflections, time of flight, etc. And these surface measurements have produced   
   some good cancer detection approaches.   
      
   So I'm looking for an "aha" moment here about some signal processing method.   
   And   
   I admit, it's unlikely to be there, since so many researchers are working in   
   this area.   
      
   But, already there are signs that you can extract very general information from   
   a human and detect very specific disease states. This flies in the face of   
   medical diagnostic thinking, which is reductionist, and not-so-much systemic.   
      
   What am I trying to do? George, mainly I like to do weak signal processing and   
   AI software. I like to think about lock-in amplifiers, signal averaging, and   
   the like. I have spent my time mainly thinking.   
      
   And I'm mainly interested in one problem: In a welter of signals, how do you   
   pull out a weak signal of interest buried in there? Is there an analogy to   
   bouncing a signal off the moon?   
      
   One way, with classic statistical training approach, is you get data on people   
   with and without TB. (just an example only.). Then you train, say with a NN,   
   to get your recognizer to indicate TB. In that way, the statistical process   
   itself can filter out the "noise."   
      
   I hope I've been able to state my case, and I feel fortunate to work with   
   brilliant people like yourself and Phil, at least in the restrictions here.   
      
   --- SoupGate-Win32 v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)