Multimodal Biomedical Imaging Systems - Biomedical Optical Imaging Technologies: Design and Applications - page 323

light applications. They have very low background and low noise, allowing for
long image acquisition time. The readout and thermal noise are minimized through
slow readout rates and cooling. Slow scan CCD cameras are generally limited in
their frame rates, and the SNR is poor when exposure times are short, unless the
specimen has extremely bright fluorescence. ICCD uses an image intensifier that
is fiber-optically coupled to the CCD chip to increase the sensitivity down to a
single photon level. The image intensifier converts incoming photons into electrons
at the photocathode, multiplies the electrons with high-voltage acceleration through
a microchannel plate, and then reconverts the multiplied electrons back to photons
through a phosphor-coated window. The photons emitted at the phosphor-coated
window are then projected onto a CCD through a fiber-optic plate. ICCD cameras
have very fast response times and the CCD camera readout is the slowest step
in the image acquisition. ICCD cameras are frequently used in studying dynamic
events and for ratio imaging of ion-sensitive fluorochromes. EMCCD is developed
for applications that demand rapid frame-rate capture at extremely low light
levels. EMCCDs employ an on-chip amplification mechanism to multiply the
photoelectrons that are generated in the silicon; therefore, the signal from a single
photon event can be amplified above the read noise floor at fast readout speeds.
A CMOS sensor is a low-cost, compact, versatile detector with the virtues of
silicon detection but without the problems of charge transfer. Rapid developments
in CMOS sensors give them a potentially important future role in fluorescence
imaging. CMOS sensors have an amplifier and digitizer associated with each
photodiode in an integrated on-chip format, which allows manipulation of indi-
vidual photodiodes, user-selectable ROI readout, high-speed sampling, electronic
shuttering, and exposure control.
The key parameters to consider when selecting the optimal detector for a specific
application include QE, internal gain, dynamic range, response speed, and noise.
The selection of detector also depends on the temporal resolution required to study
specific biological processes. For a process with picosecond or nanosecond time
resolution, for example, fluorescence decay, high-speed detectors are required.
No single detector will meet all requirements of fluorescence imaging; some
trade-offs between requirements are necessary. For example, when time is available
for image integration, a slow scan CCD usually outperforms an intensified camera in
all areas because of its higher QE and lower noise. However, when time is a critical
parameter, an intensified CCD is often the only choice. With future development,
CMOS sensors may replace the CCD in many fluorescence imaging systems.

9.3.2.6 System Considerations

The design considerations for a fluorescence imaging system include the power
and homogeneity of the excitation light, the FOV and NA of the detection system,
the sensitivity and noise characteristics of the detector, and the transmission and
blocking capabilities of the fluorescence filters.