Biomedical Engineering Reference
In-Depth Information
compatible with the emission characteristics of the indicator(s), the optical
confi guration, and the required spatial and temporal resolutions. Depending on the
staining and preparation of the tissue, optical confi guration, and imaging device
characteristics, it is currently possible to acquire images at rates of up to 1,000
frames/second (fps).
With conventional fl uorescent microscopy, a high-pressure mercury lamp or a
xenon lamp can provide stable excitation illumination over a wide range of wave-
lengths. The use of light-emitting diodes (LEDs) as illumination sources is a more
recent innovation and has been found to have several advantages over conventional
sources. An LED can supply illumination at only a single wavelength but offer a lot
of advantages as excitation sources if an appropriate wavelength is available: the
size of the source is very small, the fl uctuation in intensity due to heat production is
very low, the intensity is easily variable over a relatively wide range (reducing the
need for neutral density fi lters), and the illumination can be switched on and off
very rapidly, eliminating the need for shutters in the excitation light path.
Light scattering is one of the most serious problems that arise for in vivo brain
imaging. The scattering problem is greatly reduced through the use of confocal
laser-scanning microscopes (CLSMs), which are currently the most popular imag-
ing systems used for high-resolution Ca
2+
imaging. Inexpensive laser sources with
excellent lifetimes have recently become available. Two-photon microscopy (TPM)
has also become a leading-edge imaging technology in neuroscience. TPMs have
profound advantages for in vivo imaging because they are capable of observing
neuronal activity located at relatively deep layers in whole brain preparations
(Helmchen and Denk
2005
; Nemoto
2008
; Svoboda and Yasuda
2006
). However,
CLSM and TPM have relatively lower temporal resolution than conventional
camera-based devices for capturing images of large fi elds, because the operation of
scanning the laser over an entire fi eld of view is more time consuming than “snap-
ping” an image of the same fi eld with a conventional camera-based device. An
effective way to mitigate this limitation to the temporal resolution is to operate the
CLSM or TPM in “line-scanning mode”: the investigator captures only a linear
“transect” through the region of interest in the fi eld of view, rather than collecting
an image of the entire fi eld of view. This line-scanning mode is typically used for
measurement of Ca
2+
transients evoked by single action potentials or EPSP. In this
mode, however, it is impossible to record the two-dimensional patterns of Ca
2+
sig-
nals. Nipkow-type spinning-disk microscopes are an effective intermediate solu-
tion: they substantially improve the time resolution for confocal imaging over
conventional scanning laser confi gurations. This microscope provides a confocal
effect by spinning a disk with microlens-embedded pinholes in the excitation light
path. However, the image acquired with the Nipkow-disk microscopy is less intense
than that obtained through conventional CLSM. Recently, the use of a high-
sensitivity digital camera like the EM-CCD camera as the image acquisition device
for Nipkow-disk microscopy has enabled much brighter confocal imaging at higher
time resolution. Using this technology, investigators have detected Ca
2+
transients
evoked by individual action potentials at 150
μ
m tissue depth (Takahashi et al.
2010
;
Takahara et al.
2011
).
