Re: Improving the confocal microscope images

From: Christian Wilms (usenet01_at_out-of-phase.de)
Date: 10/27/04 

Date: Wed, 27 Oct 2004 14:22:23 +0200

You could adjust the offset value in you imaging software to "cut away"
the background. Doing so you will have a constant background with an
intensity value off zero. Of course you will be losing information if
you cut away too much.

To optimize your contrast you want to adjust your gain value inorder to
be sure you are using the full dynamic range your software has to offer.
Here you also risk losing information if you raise the gain too much as
you will saturate bright areas, losing contrast in these.

Basically one often sees "bad" confocal images in publications where the
background is all black and the bright regions are saturated. A good
image will always contain a bit of noise, as this shows that you aren't
loosing informtation in darker regions of the image.

It is very helpful to use a color look up table (LUT) which is runs from
black to white but uses a different color (e.g. blue) for zero and
another (e.g. red) for maximum (the FluoView system from Olympus has a
LUT called hi-lo for this). This simplifies setting up the offset and
gain parameters. What you will want to do is raise the offset until you
have a few blue pixels (intensity value = 0). Really only a few! And
then raise the PMT-voltage or gain-value until you have a few red pixels
(intensity = maximum). Now you have set up the system to use the full
dynamic range of your system for the given laser-intensity and specimen.

Deconvolution of a confocal image will normally "only" improve your
resolution. If the confocal is set up correctly (i.e you are using the
correct pinhole size) you shouldn't have all too much out-of-focus
signal.

You might want to grab an issue of Pawley's "Handbook of Biological
Confocal Microscopy" at your library for more background information and
a few handy tips.

h2h, Chris