Snapshot Image-Mapping Spectrometer for
Hyperspectral Fluorescence Microscopy
Liang Gao, Robert T. Kester, Nathan Hagen
and Tomasz S. Tkaczyk
We recently developed a snapshot hyperspectral imaging device
called the image-mapping spectrometer
(IMS) for dynamic cellular imaging applications. 1, 2 The IMS replaces the camera in a digital imaging system, allowing
one to add high-speed snapshot spectrum acquisition capability to a variety
of macroscopic imaging modalities—
e.g., microscopy, endoscopy, etc.—to
maximize collection speed.
The motivation to develop a snapshot hyperspectral imaging technique
comes from the high temporal resolution
requirement in time-resolved multiplexed
fluorescence imaging. 3 Conventional
instruments acquire hyperspectral datacubes (x, y, λ) through scanning, either
in the spatial domain (as in confocal laser
scanning microscopes) or in the spectral
domain (using acousto-optic or liquid-crystal tunable filters). Because scanning
instruments cannot collect light from all
elements of the dataset in parallel, there is
a loss of light throughput by a factor of N
when measuring N spectral channels.
To some extent, one could compensate for this by increasing the intensity of
illumination, such as with the high-power lasers used in confocal microscopes,
but this produces photobleaching and
photodamage to the sample. Once all of
the fluorophores have been boosted to
their excited state—a situation commonly reached by existing confocal
systems—even this method falters. For
demanding applications that require imaging of fast dynamic scenes, scanning
hyperspectral fluorescence microscopes
thus provide poor performance.
The IMS is a parallel acquisition
instrument that captures a hyperspectral datacube without scanning. It also
allows full light throughput across the
whole spectral collection range due to its
snapshot operating format. Its operation is
based on redirecting image zones through
Ly Lx Lkk-plane (a) x’ y’ 460 nm 472 nm 535 nm 538 nm 542 nm 546 nm 549 nm 520 nm 523 nm 526 nm 529 nm 463 nm 465 nm 467 nm 469 nm Elemental image 532 nm 596 nm 601 nm 605 nm 610 nm 615 nm 20mm
Assembling the full datacube by IMS and its cellular fluorescence imaging results. (a) Data produced by an image of a monochromatic spatially uniform field; 60 of them are used to produce
the full calibration dataset. (b) The acquired spectral channel images of bovine pulmonary artery
endothelial cells. Cellular nuclei labeled with DAPI are visible in the blue spectral channels; filamentous actin labeled with Alexa Fluor 488 phalloidin are seen in the green; and mitochondria
labeled with Mito Tracker Red CMXRos are visible in red.
620 nm 625 nm 630 nm 636 nm 642 nm 649 nm
the use of a custom-fabricated optical
element known as an image mapper. 4 The
image mapper is a complex custom optical
component comprised of high quality,
thin mirror facets with unique 2-D tilts.
These mirror facets reorganize the
original image onto a single large-format
CCD sensor to create optically “dark”
regions between adjacent image lines.
The full spectrum from each image line
is subsequently dispersed into the void
regions on the CCD camera. The entire
datacube (x, y, λ) is acquired instantaneously in a single integration event. This
mapping method establishes a one-to-one
correspondence between each voxel in the
datacube and pixel on the CCD camera
requiring only a simple and fast remapping algorithm for data reconstruction.
The current IMS acquires a datacube
of size 285 × 285 × 60 (x, y, λ), and a
spectral range sampled by 60 spec-
tral channels from 450 to 650 nm. For
demonstration, (b) of the figure shows 27
of the 60 total spectral channel images of
triple-labeled bovine pulmonary artery
endothelial cells, which were captured by
the IMS in a single snapshot. For future
applications, we intend on using high-
speed sCMOS detector arrays5 inside
the IMS, allowing for time-resolved fast
imaging of action potentials and full-field
Raman spectroscopy—two demanding
applications that lie at the limit of what
current instruments can measure. t
Liang Gao ( email@example.com) , Robert T. Kester, Nathan
Hagen, Tomasz S. Tkaczyk are with the department
of bioengineering, Rice University, Houston, Texas,
U.S.A. Gao and Tkaczyk are also with the Rice
Quantum Institute at Rice University. T. Tkaczyk is
also affiliated with the department of electrical and
computer engineering of Rice University.
1. L. Gao et al. Opt. Express 17, 12293-308 (2009).
2. L. Gao et al. Opt. Express 18, 14330-44 (2010).
3. D.G. Spiller et al. Nature 465 (7299), 736-45 (2010).
4. R. T. Kester et al. Appl. Opt. 49( 10), 1886–99 (2010).
5. Fairchild, Inc., Andor, Inc., and PCO, Inc., sCMOS data
50 | OPN Optics & Photonics News