Dear Group,
     There seems to be heightened interest in deconvolution
these days.
     At a recent BioRad workshop in Boston, there was some pointed
debate between proponents of confocal vs. deconvolution, and a
claim was made that deconvolution is "better" than confocal.
It seems, however, that optically rejecting the out-of-focus
light is (in general) better than mathematically rejecting it.
When pressed, the deconvolution proponent was not able
to document an instance where deconvolution exceeded confocal--
i.e. was not able to cite a reference where a sample was imaged
by BOTH methods, and the results critically compared.  I am
wondering if the group knows of any such published results?
    While deconvolution may serve as a "poor man's confocal",
especially as computing power becomes cheaper, will it ever
EQUAL confocal, much less exceed it?  Will it become
"real-time" anytime soon?
    Two other pertinent questions are: 1) is this comparison specimen-
dependent, i.e. do certain specimen characteristics make a sample
better for one method or the other? and 2) Does deconvolution improve
confocal images?  I expect that with a pinhole aperture fully closed,
deconvolution may add little, but I'd expect deconvolution to help
if one is operating with a wider-open aperture in order to decrease
illumination intensity.  Does anyone have experience with this?
    Any comments or references would be greatly appreciated.


Thanks,

Don

p.s. Thanks to everyone for the comments on laser-killing of cells.

Donald M. O'Malley
Assistant Professor
Dept. Biology
414 Mugar Hall
Northeastern University
Boston, MA 02115

phone: 617-373-2284
fax: 617-373-3724
e-mail: [log in to unmask]