Biomedical Engineering Reference
In-Depth Information
(a)
(d)
(g)
(b)
(e)
(h)
(c)
(f )
(i)
70
80
70
60
50
40
30
20
70
60
50
40
30
20
60
50
40
30
20
10
0
SHG
Picrosirius red
SHG
Picrosirius red
SHG
Picrosirius red
10
0
10
0
1
1
1
FIgurE 16.6
(
See color insert.
) Highlighted fiber orientation relative to the horizontal was measured in Gimp,
a freely available imaging processing tool. The graph shows average angle orientation with error bars indicating
(+/−) standard deviations of the 5 angle measurements per sample of both the SHG and picrosirius techniques.
Scale bars = 10 μm.
SHG image but is lost in the picrosirius red image. To quantify this further, we developed a curvelet
transform approach (described below) that can quantify the straightness/waviness of collagen fibers.
With increased collagen waviness, we would expect the standard deviation of the angular orientation to
increase, which is what we observe (data not shown).
We assume that incorporation of the picrosirius red technique into clinical settings may be more
practical than SHG since the picrosirius red technique only alters the current slide preparation protocol
and can utilize readily available hardware. The picrosirius red technique used here utilized a standard
pathology microscope and polarizers and demonstrates the ability to detect collagen and its various
signatures. However, if more robust quantitative measurements are desired, additional work with both
picrosirius red and SHG measurements is necessary. Further analysis with picrosirius red would benefit
from use of multiple polarization orientations of the light to capture all of the picric-dyed birefringent
collagen fibers, not simply the fibers in the orientation selected by the viewer, which is possible with a
pol-scope. Also, additional information is easily attainable from SHG. Using circularly polarized light
instead of linearly polarized light removes the orientation dependence of the collagen signal [44]. In
conjunction with circularly polarized light, both the forward and backward signals can be detected to
