Imaging Raman image reconstruction

Best results are obtained when spatial variation is maximized. For that reason the recovered contrast is over 140 1 when the image reconstruction is based on transmission measurements, but only about 1.4 1 for contrast based on backscattered measurements. It is likely that even better results would be obtained in both image-guided Raman spectroscopy and Raman tomography if illumination and collection fibers were placed strategically around the specimen. 

Several methods for constructing a two-dimensional Raman image from point or line spectra have already been described, and several examples are presented here. A wide variety of reconstruction methods have been reported and the examples below are certainly not comprehensive. In particular, those techniques available in commercial instruments are emphasized. Furthermore, the case of image reconstruction from an Jc-by-y array of single-point spectra is conceptually trivial and is not presented here. 

These results demonstrate that confocal Raman microspectroscopy, when coupled to image reconstruction methods such as HCA, can provide label-free methods for the visuahzation of intracellular components and processes. This subject will be elaborated further in the next section. 

Miljkovic, M., Chernenko, X, Romeo, M.J., Bird, B., Matthaus, C., and Diem, M. (2010) Label-free imaging of human cells algorithms for image reconstruction of Raman hyperspectral datasets. Analyst, 135, 2002-2013. 

Fig. 5.2. Cross section of the reconstructed image-guided Raman spectroscopic estimates of bone distribution using (a) backscattered collection using a ring/disk fiber optic probe and (b) transmission measurements using a rectangular array of collection fibers. The contrast between the bone and background from skin and tendons was more than 100-fold greater in the transmission measurements used than in the backscattered measurements (reprinted with permission from [26]. Copyright 2008 Optical Society of America)...

Fig. 6.10. In vivo multiplex CARS microspectroscopy of a NIH 3T3-L1 fibroblast cell in the high-wavenumber region where C-H stretch vibrations reside. A CARS image revealing the intracellular distribution of constituents with high densities of lipids, such as the membrane envelope of the nucleus and intracellular lipid droplet (LD) organelles. Typical MEM-reconstructed Raman spectra taken for (B) a single LD organelle that is indicated by the arrow in A, (C) the nucleus, and (D) the cytoplasm. The spectrum exposure time was 0.3 s...

Fig. 6.15. Multiplex stimulated Raman loss microspectroscopy of 20-pm polystyrene beads dispersed in water. A Raman spectra retrieved from SRL spectra of an individual bead and of bulk water at locations indicated by circles and squares, respectively, in the images. B Reconstructed Raman images of the beads representing the density maps of three characteristic Raman resonances at 1003 cm-1, 2904 cm-1, and 3066 cm-1 of polystyrene (Courtesy of Evelyn Ploetz et al., after [21])...

Raman microspectroscopy results from coupling of an optical microscope to a Raman spectrometer. The high spatial resolution of the confocal Raman microspectrometry allows the characterization of the structure of food sample at a micrometer scale. The principle of this imaging technique is based on specific vibration bands as markers of Raman technique, which permit the reconstruction of spectral images by surface scanning on an area. 

Figure 11.17. Video and Raman image of an integrated circuit, obtained with line scanning and stage translation on a Dilor XY spectrometer. Area in white box was observed with a 28 x 28 spatial grid, and the Raman image was reconstructed from 28 CCD exposures. White regions in the Raman image correspond to higher Raman intensity in the 501 to 536 cm Raman shift range.

Figure 11.22. Illustration of the use of a Powell lens to flatten the intensity profile of a Gaussian laser beam. A is the observed laser intensity along a line focus at the sample, and B is a Raman image of two 6.7 pm diameter polystyrene spheres reconstructed from a collection of line images. (Adapted from Reference 23.)...

Figure 11.23. Raman images of a pharmaceutical tablet reconstructed from 28 line images obtained with mechanical line scanning in a Dilor XY spectrometer. Both images are of the same area on the tablet, but image A is derived from an acetamidophenol band, while image B is from an excipient band. False color intensity scale is shown to the right of both images. Individual spectra are shown in Figure 11.7. (See color plates.)...

An AOTF Raman chemical imaging system has been applied to the study of polymer inclusions in a biopsied section of human breast tissue. The polymer inclusions were believed to be Dacron polyester based on the patient s medical history and previous Raman and IR data [82]. Dacron polyester patches are occasionally employed to attach breast implants to the chest wall during reconstructive and cosmetic surgery. The problems associated with silicone implants, including rupture and release of silicone gel contained in these implants, have been well publicized [83]. The goal of this study was to visualize the distribution of polyester in the thin section prepared from tissue biopsies. 

Figure 6.19 Confocal Raman image of xylem cells of wild type P. trichocarpa (a) CH intensity image (WITec 1.94 software) and (b) reconstructed image of the marked rectangle in (a) with 10 clusters. Colors in (b) are arbitrary and provide visual contrast of the...

---