Biomedical Engineering Reference
In-Depth Information
FIgurE 16.5
(
See color insert.
) Collagen in early dysplasia imaged by both polarized picrosirius and SHG
images all taken at 40× magnification from Col1a1
tm1Jae
/+ PyVT (+/+) palpable mammary mouse tissue. (a) H/E,
(b) polarized picrosirius red, (c) multiphoton and second-harmonic composite, (d) second harmonic. Zoomed in
(150%) regions demonstrate differences in collagen fiber detection between techniques. Scale bars = 10 μm.
both picrosirius red and SHG; however, a visual distinction is apparent in the signal arising from the
boundary collagen. Here, the picrosirius red signal appears saturated and fails to provide any indication
of individual wavy collagen fibers as visually seen in the SHG image. Also in Figure 16.5, the already
controversial interpretation of fiber coloration in picrosirius red is further complicated with alterations
in the acquisition settings during picrosirius red collection.
A clear distinction between picrosirius red and SHG is demonstrated in Figure 16.6, where collagen is
not readily abundant. Collagen's signal in SHG is much stronger than picrosirius red although it still is
visible in both techniques. The low collagen signal in picrosirius red may be explained by its dependency
on the picric molecule binding and aligning parallel to the collagen fiber to maximize the birefringence
of the dyed collagen fiber. SHG is an intrinsic nonlinear optical effect and is sensitive to both the con-
centration of collagen (squared) and collagen organization. Due to this fundamental limitation, it is
difficult to perform a robust quantitative analysis comparing picrosirius red and SHG despite previous
attempts.
We attempted to show the ability of both techniques to decipher small angular changes in collagen
fibers in the tumor stroma boundary by selecting five specific fibers in each picrosirius red and SHG
image and measuring its angular orientation with respect to the horizontal by hand. Superimposed lines
on the collagen were best matched to the overall shape of the collagen fibers to permit relative comparison
between the measurements of both techniques (Figure 16.6). Direct comparison between the angular
measurements is impossible due to the following limitations: (1) the image frame of the picrosirius red
and SHG have a rotation; and (2) the areas of interest are different pathological 5 μm sections. Relative to
each other, there is no quantitative difference shown between average angular measurements or standard
deviation measurements. Therefore, with limited analysis tools, it seems as if picrosirius red and SHG are
equivalent in their ability to detect collagen signals in areas of abundant collagen. However, we anticipate
seeing a difference in angular orientation sensitivity with more robust analysis tools.
Upon further digital zoom, as shown in panels (a) and (b) of Figure 16.5, we note that the SHG
images show finer details in the structure of collagen. The waviness of the collagen is still intact in the
