|
Signal-to-Noise Ratio Comparison of CardioCMOS
and CardioCCD
In terms of signal-to-noise ratio the best a camera
system can do is to approach the shot-noise limit. Some of the factors
in the camera design affect how close the camera approaches shot-noise
performance and others affect the range of light intensities over
which this performance is achieved. In addition, extraneous factors
can make the performance worse than the shot-noise limit.
Shot Noise
The limit of accuracy with which light can
be measured is set by the shot noise arising from the statistical
nature of photon emission and detection. Fluctuations in the number
of photons emitted per unit time will occur, and, if an ideal light
source emits an average of N photons/ms, the root- mean-square deviation
in the number emitted is the square root of N. The effects of this
relationship are indicated by the green line in the Figure which
plots the light intensity divided by the noise versus the number
of photons measured per ms. At high intensities this ratio is large
and thus small changes in intensity can be detected. For example,
at 108 photons/ms a fractional intensity change of 1.0%
can be measured with a signal-to-noise ratio of 100. On the other
hand, at low intensities this ratio of intensity divided by noise
is small and only large signals can be detected. For example, at
104 photons/msec the same fractional change of 1.0% can be measured
with a signal-to-noise ratio of 1 only after averaging 100 trials.
The figure also indicates the performance
of the two RedShirtImaging camera systems, the CCD camera system (RedShirtImagingCCD;
blue lines) and the CMOS camera
(RedShirtImagingCMOS, green lines
). RedShirtImagingCMOS approaches the shot-noise
limitation over the range of intensities from 105 to 108 photons/ms.
This is the range of intensities obtained in absorption measurements and fluorescence measurements
on intact hearts. On the other hand,RedShirtImagingCCD approaches the shot noise limit over the range
of intensities from 102 to 5x105 photons/ms. This is the range of intensities
obtained from fluorescence experiments on individual cells.
The RedShirtImagingCMOS camera doesn't get quite as close to the ideal as the RedShirtImagingCCD
camera because its quantum efficiency and fill factor are not as good.
Saturation
The high intensity limit of the RedShirtImagingCCD
is set by the light intensity which fills the electron wells on
the CCD chip. This accounts for the bending over of the camera performance
at high light intensity. Even though the RedShirtImagingCCD camera has
a large well-size compared to other CCD cameras, it will not be
optimal for measurements of absorption. The light intensity would
have to be reduced with a consequent decrease in signal-to-noise
ratio.
Dark Noise
Dark noise will degrade the signal-to-noise ratio at low light levels
and accounts for the bending down of the camera curves away from ideal at lower light intensity.
The RedShirtImagingCCD is a cooled CCD camera and its dark noise is much lower than that of the
RedShirtImagingCMOS or other fast CMOS cameras. The excess dark noise in RedShirtImagingCMOS
accounts for the fact that its low intensity segment in the Figure is substantially to the right
of the low intensity segment for the RedShirtImagingCCD. The dark noise of the RedShirtImagingCCD
is lower than other commercially available CCD's in its performance range.
The RedShirtImagingCMOS camera has very large wells and is unlikely to saturate in biological
experiments that use tungsten filament or arc lamp light sources.
Extraneous Noise
A second type of noise, termed extraneous or technical noise, is often
detected at higher light intensities where the sensitivity of the measurement is high because
the fractional shot noise is low. There are several sources of extraneous noise. One type is
caused by fluctuations in the output of the light source. Other sources are vibrations and movement
of the preparation. Extraneous noise is likely to limit the signal to noise ratio at intensities
higher than 108 photons/msec/pixel.
|
