Biomedical Engineering Reference
In-Depth Information
Raw
Hologram
Lensfree
Image
¥
Microscope
10
b
a
400k
±
5% line
Reconstruction
Raw Hologram
y = x
350k
-4.57%
300k
Rec.
250k
-3.33%
Rec.
+3.15%
Rec.
200k
50 100
200 300
[Volume: fL]
1200
Rec.
150k
900
-0.41%
Coulter Result
Reconstructed
Hologram
Results
N
rec.
= 17,684
Rec.
100k
600
+2.25%
-3.79%
Raw
300
50k
[Area: Pixel]
-2.25%
Raw
0
0
50
100
150
0
0
50k
200k
Manual Microscope Count [cells/µL]
100k
150k
250k
300k
350k
400k
Fig. 4.4
(
a
) Automated counting of RBCs in the hologram domain (
pink curve with square
markers
) and the reconstructed image domain (
blue curve with triangular markers
) show that cells
can be accurately counted with <5 % error up to densities of
0:4 Million cells/L.
Inset
shows
the calculated volume histogram of RBCs compared against a Coulter counter measurement of
the same blood sample. (
b
) Measured holograms, reconstructed lensfree images and conventional
microscope (10
, 0.25-NA) images are provided for cropped regions in the FOV at three different
cell density levels
from a much larger FOV of 24 mm
2
across which the imaging quality does not
noticeably vary, enabling rapid wide-field imaging with a single-shot holographic
image.
Despite their simplicity, lensfree telemedicine microscopes can perform
advanced tasks for point-of-care diagnostics such as automated blood analysis.
Toward this end, we recently demonstrated cytometry on a chip using this lensfree
imaging modality [
12
]. Owing to the large FOV of our platform, large numbers
of cells can be accurately counted in less than 1 s to measure the density of
different cell types in whole blood. To achieve that, either the cell holograms can be
enumerated by automated pattern matching using a library of cell holograms, or the
reconstructed cell images can be counted. The advantage of the latter is that larger
numbers of cells can be accurately counted in the image domain as cell images
do not overlap at high densities as much as their diffraction holograms do (see
Fig.
4.4
). As a result, RBCs can be counted with <5 % error up to densities reaching
0.4 million cells/L. In addition to counting, volume histograms of RBCs can also
be calculated as validated against a commercial Coulter counter (see the inset in
the plot shown in Fig.
4.4
). The phase imaging capability of lensfree holographic
microscope plays a key role for this task, as the reconstructed phase information
is used to estimate the volume of each individual cell. The same platform can
also be utilized to determine the hemoglobin concentration within whole blood
by measuring the photon transmission through a cuvette filled with whole blood
(see Fig.
4.5
a). Additionally, WBCs can also be counted with less than 10 % error
