Molecular Devices ImageXpress Confocal HT.ai, Руководство пользователя

Страница: 60 / 256

ImageXpress Confocal HT.ai High-Content Imaging System User Guide
60 5084773 A
Illumination Settings
You select the illumination setting for your assay. The Semrock website at www.semrock.com
is a good resource to help you determine the best filters to use with your fluorophores. The
ImageXpress Confocal HT.ai system uses a standard, selectable light source filter to create the
correct excitation wavelength, and a filter wheel to select the corresponding emission
wavelength filtration for the experiment. It uses dichroic mirrors to separate the excitation
wavelength from the emission wavelength. Specific filter sets are designed for use with
specific stains. See
Table 8-1: Settings for Common Fluorophores on page 208 .
Exposure Time
An appropriate exposure time for each wavelength is crucial for acquisition and analysis. Click
Auto Expose on one of the Wavelength tabs in the Plate Acquisition Setup dialog to provide a
good starting point and then adjust the exposure time so that the grayscale intensity within a
cell is at least three times the intensity of the background. If the background comes from out-
of-plane fluorescence, consider using confocal imaging to improve the image quality. If you
expect the intensity to vary with the phenotype, adjust the exposure time with both positive
and negative controls so that the bright samples are not saturating and the dim samples are still
visible.
Signal-to-Noise (S/N) is the ratio of the signal from the foreground objects over the noise from
the background variation, signal variation, and other forms of noise from the system. S/N is an
important leading indicator of how difficult it will be for the software to accurately discriminate
important features in an image from the background. If the S/N is large (for example, >10), then
there is high confidence that the software can distinguish foreground pixels from background
pixels in the image. If the S/N is relatively small (for example, <2), then it is more difficult for the
software to distinguish between objects in the foreground and background for accurate count,
morphology, area, signal intensity, and other measurements.
To increase S/N, consider using one of the following methods:
Increase the image exposure. Longer exposures provide higher signal and higher
background in an image, so changing the exposure alone might not increase S/N as
expected and should be carefully measured before using long exposures (>1 second).
Longer exposures can cause photo bleaching damage and saturate the camera. Intensity
measurements of an overexposed image are not accurate, and you should avoid
generating these images. One exception to this rule is when you are interested in
extremely faint features of the sample that are otherwise not visible. A good example of
this is overexposure of a neurite image where you are not interested in the bright cell
body, but you are interested in the weakly stained outgrowths. Longer exposure times
also increase the overall plate acquisition time.
Decrease background variation by lowering background levels from the sample, system,
or other sources. This could be achieved by changing to confocal imaging with longer
exposures, lowering the sample background, or changing the consumable type.
Average images. In the composite of averaged images, random noise in the background
should follow the central limit theorem and therefore lower pixel-to-pixel variation in object
foreground and background pixels. With smoother background variation, the S/N will
increase. The number of images to be averaged will depend on the sources of variation
and stability of the sample (in other words, the amount of photobleaching or
photodamage).