Page 286 - Optofluidics Fundamentals, Devices, and Applications

260 Cha pte r Ele v e n

the quality of clinical care by forming the imaging engine of cheap
point-of-care analysis units and by cutting down on contamination
risks (by being disposable).
The field of optofluidics offers us an opportunity to redesign the
conventional microscopy from the ground up. The optofluidic
microscope (OFM) developed by our group capitalizes on the ease
of transporting cells and microorganisms via microfluidic flow and
the high-quality optical sensor grid that are readily available at
remarkably low cost [2].
This chapter is divided into four sections. In Sec. 11-2, we will
introduce the OFM’s operating principle. In Sec. 11-3, we will sum-
marize the experimental implementation and evaluation of the first
OFM prototypes. In Sec. 11-4, we will discuss some of the applica-
tions that the OFM is well suited to make an impact.

11-2 Operating Principle
The OFM principle is best explained by recounting the phenomenon
that inspired the idea—the “floater” phenomenon that most of us occa-
sionally observe when looking at a clear patch of sky. Floaters are
caused by debris in the vitreous humor that drifts close to the retina.
Under uniform illumination, they cast sharp shadows onto the retina
and “appear in our perception.” The clarity of floaters is a direct func-
tion of their proximity to the retina; the closer they are, the sharper the
shadows cast. Despite the fact that floaters are tiny, we often see them
with excellent detail. It is also interesting to note that our ability to see
these tiny objects is not influenced by our eye glasses or the intrinsic
lenses in our eyes (if in doubt, try putting on or off a pair of glasses the
next time you see floaters). This observation points to the fact a direct
projection imaging strategy (basis of the floater phenomenon) is capa-
ble of rendering high-resolution images as long as (1) we can place the
target close to the sensor grid, and (2) the sensor grid pixels are small.
The direct projection imaging strategy has previously been used
by other groups [3]; however, the quality of the images is less than
satisfactory for microscopy applications as the image resolution is
bounded by the size of the sensor pixel. Since the typical pixel size of
a commercial CCD or CMOS sensor is larger than 3 μm (getting down
to smaller pixel size is difficult from a semiconductor fabrication
point of view), the resolution achievable is much poorer than the res-
olution achieved with a conventional microscope.
It is difficult to imagine that a single-time-point direct projection
imaging strategy for collecting images at resolution better than the
sensor pixel size exists. However, if we permit ourselves to exploit the
time dimension during the image-acquisition process, it is possible to
develop viable high-resolution direct projection imaging strategies in
which resolution and sensor pixel size are independent. To begin,
consider the following sensing platform—a sensor grid that is coated