Page 186 - Vibrational Spectroscopic Imaging for Biomedical Applications

162 Cha pte r S i x

With the advent of improved spectrometers, Raman microscopes,
and faster CCDs, Raman spectroscopy of cells has improved greatly.
Uzunbajakava et al. showed Raman mapped images of protein, DNA,
RNA, and lipid distribution in live and apoptotic cells, and determined
1,2
the effect of high laser powers on their samples. In the same year
Krafft and coworkers observed in their mapped Raman images nucleus,
organelles, and membranes of freeze-dried and living cells. They found
that the quality of the images was influenced by the hydration state of
3
the molecules in the cells. In 2005, Creely et al. utilized optical trapping
technology to collect Raman mapped images of living cells in suspen-
4
sion. The research group of Diem demonstrated label-free detection of
drug uptake in cells with Raman mapped images. By staining the
Raman mapped cell after acquisition, they were able to prove that mito-
chondria-rich regions were correctly identified in the Raman image. 52
Recently, Raman mapping and imaging have been employed in
conjunction with labeling or signal enhancement techniques. In work
reported by Keren et al., SERS nanoparticles were injected into a live
mouse. These were observed in the mouse’s liver by collecting SERS
5
mapped images through the mouse’s skin. Chao et al. used nanometer-
sized diamond particles as detection probes for biolabeling human lung
53
epithelial cells at specific interaction sites. Raman point mapping of
labeled cells revealed an easy method for monitoring cell interactions.
The widefield Raman imaging literature is dominated by Treado
and coworkers, who were first to demonstrate a novel method for
measuring widefield Raman images using an AOTF for wavelength
31
selection. In 1996, Schaeberle et al. demonstrated the first applica-
tion of widefield Raman imaging (called “Raman chemical imaging”
by the group) to pathology by targeting polymer inclusions some-
times found in augmented breast tissues. That same year, Morris et al.
introduced the LCTF as an alternative method for collecting wide-
32
field Raman images of biological materials. Kline and Treado
employed LCTF technology to report Raman images of chicken breast
35
tissue. They demonstrated the capability of this technology to high-
light lipid and protein distribution within the tissue.
In more recent tissue work, the results of a preliminary study
indicate that widefield Raman imaging can differentiate between
two types of kidney cancer commonly difficult to diagnose by
46
pathologists. In this investigation, Raman images, collected from
identifiable kidney cancers, were used to create a classification model.
The generated model was then utilized to classify image data collected
from test biopsy samples. The model exhibited 79 percent sensitivity
and 85 percent specificity for the relevant kidney cancers.
Widefield Raman imaging has also been employed at a cellular level.
Maier and Treado reported three-dimensional widefield Raman imaging
54
of a single human cell. They discussed the use of deconvolution and
show the improvements to the image as a result of the methodology.
Panza and Maier demonstrated the employment of widefield Raman