A Multimode nonlinear optical imaging technique based on the combination of multichannel mode and Lambda mode is developed to investigate human dermis. Our findings show that this technique not only improves the image contrast of the structural proteins of extracellular matrix (ECM) but also provides an image-guided spectral analysis method to identify both cellular and ECM intrinsic components including collagen, elastin, NAD(P)H and flavin. By the combined use of multichannel mode and Lambda mode in tandem, the obtained in-depth two photon-excited fluorescence (TPEF) and second-harmonic generation (SHG) imaging and TPEF/SHG signals depth-dependence decay can offer a sensitive tool for obtaining quantitative tissue structural and biochemical information. These results suggest that the technique has the potential to provide more accurate information for determining tissue physiological and pathological states.
©2006 Optical Society of America
An emerging application of multiphoton microscopy (MPM) is the observation of unstained samples based on intrinsic sources of nonlinear signals [1, 2] such as two photon-excited fluorescence (TPEF) and second-harmonic generation (SHG). It can provide high contrast and optical sectioning capabilities for involving nonlinear light-matter interactions. The ability of MPM to produce images deep in optically thick preparations is crucial for intravital tissue microscopy. Moreover, SHG enables direct imaging of anisotropic biological structures, such as collagen [3–5]. SHG can produce high-resolution, high contrast images of tissue morphology [2, 6] and has also be applied to study dynamics in tissue physiology, such as monitoring collagen modification in tumor growth , and optically recording action potential change in neuron cells . Recently, because these imaging (TPEF and SHG) modalities involve different contrast mechanisms and can provide in tandem complementary information about tissue structure and function, the combination of TPEF and SHG has attracted more attention. These imaging modalities are easy to implement simultaneously and differ only in optical filter selection and detector placement. The combination of TPEF and SHG has been implemented for the study of cells [9, 10], thin tissue sections , and for the more practical case of thick, unstained living specimens [6, 12].
In addition to structural imaging, spectral analysis can provide complementary functional information. Spectral analysis is a powerful method to identify intrinsic biochemical species. The relative abundance of intrinsic components is related to tissue physiological and pathological states . Spectral analysis has been used to characterize tissue type, monitor the physiological and pathological states of tissue [13–16]. However, the traditional tissue spectroscopy mostly bases on the overall spectrum resulting from a mixture of many different intrinsic components. It is apparent that the interpretation of the bulk spectrum is difficult, and it is nearly impossible to identify all the chemical components present within the tissue or determine their concentration. This fact prompts further investigation into the spectra and distribution of each biochemical species. By linking the spectroscopic information to tissue morphological structures, it is possible to study the spectral properties and distributions of each biochemical species . The method is significant for tissue biochemical analysis.
Collagen and elastin are important structural proteins of extracellular matrix (ECM). Collagen plays an important structural role in skin, interstitial tissue and basal laminae. Modifications of the collagen fibrillar matrix structure are associated with various physiologic processes, such as wound healing, photoaging, cornea diseases, osteoarthritis, liver fibrosis, and cancer [4, 18–22]. Elastin is the protein responsible for the characteristic elastic properties of many tissues [23–25]. The elastic function complements collagen fibrils, which impart tensile strength. However, to understand the healthy and diseased tissues, a complete understanding of the arrangement and modification of the major structural proteins, such as collagen and elastin, is crucial. Thus, the visualization of the major structural proteins of the extracellular matrix (ECM), such as collagen and elastin, is of great value in gaining structural and diagnostic information. Cellular NAD(P)H and flavin carry information of cellular metabolism. It is reported in previous work that the ratio of NAD(P)H over flavin fluorescence, a parameter related to the tissue pathology [26–28]. The investigation of NAD(P)H and flavin fluorescence is of significance for evaluating the redox ratio and monitoring the metabolic state in tissue. To date, few techniques can accomplish the selective visualization of the major structural proteins of ECM and the identification of tissue intrinsic components.
In this study, we demonstrate for the first time the multimode nonlinear imaging technique to carry out the selective visualization of ECM and the identification of dermis intrinsic components by combining multichannel mode with Lambda mode. The technique is illustrated by imaging human dermis using a mode-locked femtosecond Ti: sapphire laser coupled to a Zeiss LSM 510 META laser scanning microscopy. We also show that the technique can provide some quantitative tissue structural and biochemical information. Furthermore, we qualitatively interpret the origin of the backward SHG signals from tissue.
2. Materials and methods
All the experiments described in this paper were based on ex-vivo human skin. Human skin was obtained from Affiliated Xiehe Hospital of Fujian Medical University. The specimens were stored at a bottle of liquid nitrogen (-196°C) before they were used. The skin samples were excided perpendicular to the epidermal layer so that each section comprised a complete transverse cross-section of the epidermal and dermal layers and sandwiched between the microscope slide and a piece of the cover glass. To avoid dehydration or shrinkage during the whole imaging process, the specimen was sprinkled with PBS solution (PH 7.4). The dermal part of ex-vivo human skin is our region of interest.
In this study, the multimode nonlinear optical imaging based on the combination of multichannel mode and Lambda mode was performed on a Zeiss LSM 510 META laser scanning microscopy equipped with a mode-locked femtosecond Ti: sapphire laser (110fs, 76MHz), tunable from 700nm to 980nm (Coherent Mira 900-F), as shown in Fig. 1.
The polarization direction of the laser light is the horizontal polarization. An Acousto-Optic Modulator (AOM) is used to control the laser intensity attenuation. A Plan-Apochromat 63× (N.A.=1.4) oil immersion objective (Zeiss) was employed for focusing the excitation beam and for collecting of the backward signals. We detected all signals in the backward direction. The signals were directed by a main dichroic beam splitter (MDBS) to the META detector. The META detector with 32-gated photon counting module is used to collect x-y images at a series of emission wavelengths, covering from 377nm to 716nm. Each photon counting module can cover a spectral range of 10.7nm. An IR beam block filters ( Zeiss KP650), which is in front of META detector, was be used to ensure that excitation light was filtered out and only emission signals were recorded. The LSM 510 META laser scanning microscopy is controlled via a standard high-end Pentium PC and linked to the electronic control system via an ultrafast SCSi interface. This system has two especial imaging modes: multichannel mode and Lambda mode. These two modes use the same detector that is META detector to acquire the imaging. The multichannel mode has eight independent-channels and each channel can selectively be set to detect emission signals within the random range from 377 to 716nm to achieve imaging. The Lambda mode permits to simultaneously record the spectral resolved image and the corresponding spectra by creating emission lambda stacks. In this study, we have combined multichannel mode with Lambda mode to carry out multimode nonlinear imaging. In multichannel mode, we selected two independent-channels from eight channels to achieve SHG (387–409nm) and TPEF (447–628nm) imaging of human dermis, respectively. The SHG/TPEF images were subsequently recorded and generally attained within a few seconds. In our experiments, if the two channels (SHG and TPEF) were recorded simultaneously, which they bear the same system parameters, we found that the integrated total intensity of SHG is about 1/5 of that of TPEF (as indicated in the Fig. 5 b). Thus, when SHG imaging is on optimal, it is more likely that the much higher relative TPEF intensity obscure the SHG signal; when TPEF imaging is on optimal, the SHG signal will become much weaker. The acquired SHG/TPEF image bears low contrast and signal-to-background. So, in this study we used two independent channels and they can acquire images with the optimal system parameters, including the laser intensity, the detection gain, amplifier offset, and amplifier gain, which can enhance the image contrast. This is an advantage of the multichannel mode. At the same time, in Lambda mode we acquire the spectral resolved TPEF/SHG image and the corresponding spectra in the same region as using the multichannel mode. Using our setting, about 1s of exposure were necessary for the alternation of two modes. The acquisition of a single 512×512 pixels image was generally achieved within a few seconds. But, note that all the images presented in this article are not 512×512 pixels in size because of taking into account the size of images and discussed contents. The average laser power at the specimen was maintained at <6mW and no photobleaching was observed at this low power level.
3. Results and discussion
3.1. Multimode nonlinear optical imaging of human skin
In order to demonstrate multimode nonlinear imaging technique, we have examined samples of human skin. Human skin is comprised of two layers: the epidermis and the dermis. The epidermis is the outermost portion of the skin and is a stratified squamous epithelium composed mainly of keratinocytes. The dermis provides structural support for the epidermis and is primarily composed of a collagen and elastin matrix with a sparse population of cells such as fibroblasts. In this study, we mainly focus on the dermis. Figure 2 shows a TPEF/SHG image of human dermis using the multichannel mode. As can be seen from Fig. 2 a, the collagen mesh of dermis can be imaged alone by one channel (387–409nm, green color-coded). Similarly, in the Fig. 2 b the elastin and fibroblasts components of dermis can be isolated from collagen using another channel (447–628nm, red color-coded). Overlaying two channels yields a high-contrast TPEF/SHG image of human dermis, where structural details of the sample can be readily distinguished, as shown in Fig. 2 c. The image (green color-coded) shows a fine mesh morphology which corresponds to collagen and probes the distribution and orientation of collagen. Collagen is the predominant structural protein in most biological tissues, as well as an efficient source of SHG [3–5]. Modifications of the collagen fibrillar matrix structure are associated with various physiologic processes, such as wound healing, aging, diabetes, and cancer. Therefore, SHG is very promising as a sensitive probe in tissue morphology and physiology studies. The image (red color-coded) displays the morphology of thick ropes that corresponds to elastin fibers with diameters of ≈ 1.94μm. Elastin that coil and recoils like a spring within the elastic fibers of connective tissue and accounts for the elasticity of structures such the skin, blood vessels, heart, lungs, intestines, tendons, and ligaments. However, no cells can be visualized in the dermis. This may be responsible for two factors. It is possible that the cells are relatively sparse. It is more likely that the much higher relative fluorescence intensity of elastic fibers obscure the much dimmer fluorescence from the fibroblasts. Note that the image contrast is significantly improved in the present study as compared with the previous study [12, 25]. Since two channels are independent and can possess different system parameters as shown in Table 1, they can acquire images with the optimal system parameters which can enhance the image contrast. The method is effective in improving the image (SHG and TPEF) contrast.
Multichannel mode has also capability to generate high-contrast structural images deep inside highly scattering tissues. It was found that detailed, high-contrast features could be resolved in TPEF/SHG optical sections throughout the full ~ 60μm thickness of a freshly dissected, unfixed sample of human dermis (Fig. 3 a). Figures 3 b–h shows TPEF/SHG images from human dermis at depths of 0, 10, 20, 30, 40, 50, 60 μm, for λex =800nm. As represented in Fig. 3 b–h, the distribution and content of ECM components are visible. Optical sections at various depths can better reveal fine details of the internal tissue structure. The signal progressively degrades with increasing depth, but the TPEF/SHG images still bear high-contrast. However, the maximum imaging depth reached here is only limited by the working distance of microscopy objective and the low value of excitation power used (PSHG = 6.0mw, PTPEF = 2.6mw). Recently, some groups have reported the maximum depth of two-photon imaging can reach up to one millimeter , while leaving the tissue intact. Further improving our system may permit deeper imaging.
At the same time, in the Lambda-mode we obtain the spectral resolved TPEF/SHG image of human dermis from the same region as using the multichannel mode and the corresponding spectra from regions of interest (ROI), as shown in the Fig. 4. The spectral resolved TPEF/SHG image could provide visual identification between the different tissue structures attributed to the differences in their biochemical components. It was obvious that in the Fig. 4 there are two main color structures: blue and purple. The blue-fluorescing rope-like structures are likely to be elastin. The purple color refers to the second-harmonic signal generated by collagen. Thus, we can select our ROI, including certain intrinsic component, based on the structural information. Then the emission spectra of different ROI can be obtained by taking the image of ROI and averaging the intensity from each pixel with this region. In the Fig. 4, we obtained the spectra from our ROI and each spectrum is normalized to the maximal peak intensity. In the ROI 1, the emission spectra reveals strong SHG signals manifested by a narrow peak at half the excitation and a bandwidth (full width a half-maximum) in accordance with the excitation laser spectral width. Moreover, the signal intensity exhibited a quadratic dependence on the laser power. Collagen is well known to produce SHG and is responsible for this signal [3, 5]. However, almost no TPEF signal was observed. The result was in accordance with previous study, which no TPEF signal from collage was detected for excitation wavelength of 800nm .
In the ROI 2, there are two distinct peaks at 470nm and 500nm. It was reported in previous work that cellular contributions (NAD(P)H and flavin) and elastin emit this band [13–15, 17, 26–28]. Thus, we believe that the corresponding to NAD(P)H and flavin in the fibroblasts, and elastin contribute to two peaks, respectively. Specifically, the 470nm peak corresponds to the combination of NAD(P)H and flavin fluorescence. The 500nm peak is associated with the presence of elastin. ROI 3 displays the overall spectrum from NAD(P)H and flavin in the fibroblasts, collagen, and elastin. It is apparent that this spectral analysis method can better identify tissue intrinsic components to analysis different components spectra. It is so-called the image-guided spectral analysis method. In addition, because cellular NAD(P)H and flavin not only carry information of cellular metabolism but also relate to the tissue pathology, the method can be used for evaluating the redox ratio and monitoring the metabolic state in tissue.
In the Lambda-mode, we could also investigate the dependence of the SHG and TPEF signals intensity on imaging depth in turbid tissues. The spectral resolved TPEF/SHG images from human dermis at various depths were acquired. The emission spectra were obtained by taking the image of the whole optical section plane and averaging the intensity from each pixel within the image. To suppress the 470 nm autofluorescence peak of NAD(P)H, 830 nm excitation wavelength was used instead of 800 nm. Figure 5 (a) shows the emission spectra from human dermis at depths 0, 20, 40, 60 μm, while Fig. 5 (b) gives integrated total intensity of SHG and TPEF spectra versus penetration depth, z, for λex = 830nm. To ensure the reproducibility of the results, the experiment was repeated three times. The SHG and TPEF signals were observed to decay exponentially with depth. Multiple light scattering decreases the number of excitation photons reaching the focus plane. Multiple light scattering also decreases the number of SHG and TPEF photons collected by the objective lens. Thus, the SHG and TPEF signals decay with depth. The detected signal intensity, I, decays as a function of depth, z, according to:
where A is an attenuation coefficient that is a function of the sample absorption and scattering properties at the both the excitation and emission wavelengths. The inverse of A yields an attenuation length, latt . Figure 5 (c) and (d) show the plot of the natural logarithm of SHG and TPEF integrated total intensity versus depth, z, with and ≈ 25.6μm. The SHG (purple light) wavelength is shorter than that of TPEF (blue light), the sample absorption and scattering properties of SHG is bigger [30, 31]. So, the attenuation of SHG is more significant than TPEF, which is consistent with the above results. In addition, it is well known that refractive index mismatch induced spherical aberration is a ubiquitous problem for in-depth optical biological imaging. However, recently Dong’s group has used it to determine the refractive indices of uniformly luminescent specimens by measuring the signals decay profile as a function of imaging depth . In other words, the signals decay profile as a function of imaging depth can reflect changes of the refractive indices of specimens. Here, the SHG and TPEF signals decay profile as a function of imaging depth also can use to monitor changes of the refractive indices of biological tissue. These finding suggest that the SHG and TPEF depth-dependent decay can provide a general estimate of optical properties, which are critical parameter for deep-tissue imaging and reflect various physiologic processes . The SHG and TPEF depth-dependent decay may have the potential to be as sensitive tools for obtaining quantitative tissue information.
To sum up, the multimode nonlinear optical imaging technique can allow direct visualization of ECM components and the identification of both cellular and ECM intrinsic components. The combined use of multichannel mode and Lambda mode in tandem can obtain quantitative information regarding the biomorphology and biochemistry of tissue.
3.2. Origin of the backward SHG signal
Our study collects SHG signals in a backward geometry and the excitation light power is relatively lower. Because SHG is the coherent process and the majority of the second-harmonic wave copropagates with the excitation laser beam, it is necessary to understand the origin of the backward SHG signal from the sample. Three main sources, we believe, contribute to the backward SHG signals. Firstly, the backward SHG signals are directly emitted from the dermis collagen fibril. Emitted SHG is generally anisotropic because phase matching constraints of coherent scattering. As indicated by recent theoretical and experimental works[5, 34], objects with the axial size on the order of the second harmonic wavelength exhibits forward directed SHG, while objects with a axial size less than λ/10 (approx. 40nm)are estimated to produce nearly equal backward and forward SHG signals. As proposed by the previous investigation of tendon collagen fibrils , the dermis collagen fibrils seem possess inhomogeneous tubelike structures. The collagen microfibrils might be randomly arranged in the core but are well aligned in the shell (thickness < 40nm) the dermis fibril, generating evenly distributed forward/backward SHG. Secondly, because human dermis is opaque, the backward SHG signals may be attributed to the backscattering of the forward-generated SHG signals. Thirdly, the backscattering of ballistic excitation photons as the excitation light induces the ‘forward SHG signals’, that is, the backward SHG signals. The existence of backward SHG signals in tissue is fortuitous for in vivo tissue imaging because collecting forward propagating signals is impractical in many cases.
We have developed a multimode nonlinear optical imaging technique based on the combination of multichannel mode and Lambda mode and applied it to image human skin. The experimental results demonstrates that the technique can permit high-contrast visualization of ECM intrinsic components, and identify both cellular and ECM intrinsic components including collagen, elastin, NAD(P)H and flavin. Our findings indicate that in-depth TPEF/SHG imaging and TPEF/SHG signals depth-dependence decay provide a means for extracting quantitative structural and biochemical information from tissue by the combined use of multichannel mode and Lambda mode in tandem.
A major advantage of the technique can provide unique contrast enhancement to the previous study in TPEF/SHG image by using two-independent channels with the optimal system parameters. Another major advantage is that it is a noninvasive technique that relies exclusively on intrinsic signals, and does not required extrinsic probes, which can change the physiologic state of the tissue. Third, the technique can offer an image-guided spectral analysis method to identify tissue biochemical components.
Until now, few other techniques can provide a direct visualization of ECM intrinsic components, and identify both cellular and ECM intrinsic components. The technique reported here opens many possibilities of studying physiological activity of intact tissues. The technique provides a new tool to map the distribution of ECM components and monitor cellular redox state in intact tissue, providing quantitative information regarding biomorphology and biochemistry of tissue, and may prove useful to clinical diagnostics as well as to basic biological research.
The project was supported by the National Natural Science Foundation of China (No. 60508017), Program for New Century Excellent Talents in Fujian Province University, the Natural Science Foundation of Fujian Province of China (No. A0510015), the Innovation Foundation for Young Technological Talents of Fujian Province (NO.2003J027), and the Scientific Program of the Educational Hall of Fujian Province (NO.JA05215).
References and links
2. W.R. Zipfel, R.M. Williams, R. Christie, A.Y. Nikitin, B.T. Hyman, and W.W. Webb, “Live tissue intrinsic emission microscopy using multiphoton-excited native fluorescence and second harmonic generation,” Proc. Natl. Acad. Sci. USA. 100, 7075–7080 (2003). [CrossRef] [PubMed]
5. R.M. Williams, W.R. Zipfel, and W.W. Webb, “Interpreting second harmonic generation images of collagen I fibrils,” Biophys. J. 88, 1377–1386 (2005). [CrossRef]
6. A. Zoumi, A. Yeh, and B.J. Tromberg, “Imaging cells and extracellular matrix in vivo by using second-harmonic generation and two-photon excited fluorescence,” Proc. Natl. Acad. Sci. USA. 99, 11014–11019 (2002). [CrossRef] [PubMed]
7. E. Brown, T McKee, E. diTomaso, A. Pluen, B. Seed, Y. Boucher, and R.K. Jain, “Dynamic imaging of collagen and its modulation in tumors in vivo using second-harmonic generation,” Nat. Med. 9, 796–800 (2003). [CrossRef] [PubMed]
10. P.J. Campagnola, H.A. Clark, W.A. Mohler, A. Lewis, and L.M. Loew, “Second-harmonic imaging microscopy of living cells,” J. Biomed. Opt. 6, 227–286 (2001). [CrossRef]
11. P.J. Campagnola, A.C. Millard, M. Terasaki, P.E. Hoppe, C.J. Malone, and W.A. Mohle, “Three-dimensional high-resolution second-harmonic generation imaging of endogenous structural proteins in biological tissues,” Biophys. J. 81, 493–508 (2002). [CrossRef]
12. Y.C. Guo, H.E. Savage, F. Liu, S.P. Schantz, P.P. Ho, and R.R. Alfano, “Subsurface tumor progression investigated by noninvasive optical second harmonic tomography,” Proc. Natl. Acad. Sci. USA. 96, 10854–10856 (1999). [CrossRef] [PubMed]
13. R.K. Rebecca and S.M. Eva, “Quantitative optical spectroscopy for tissue diagnosis,” Annu. Rev. Phys. Chem. 47, 555–606 (1996). [CrossRef]
14. I. Georgakoudi, B.C. Jacobson, M.G. Muller, E.E. Sheets, K. Badizadegan, D.L. C. Locke, C.P. Crum, C.W. Boone, R. R. Dasari, J.V. Dam, and M.S. Feld, “NAD(P)H and collagen as in vivo quantitative fluorescent biomarkers of epithelial precancerous changes,” Cancer Res. 62, 682–687(2002). [PubMed]
16. M.C. Skala, J.M. Squirrell, K.M. Vrotsos, J.C. Eickhoff, A. Gendron-Fitzpatrick, K.W. Eliceiri, and N. Ramanujam, “Multiphoton microscopy of endogenous fluorescence differentiates normal, precancerous, and cancerous squamous,” Cancer. Res. 65, 1180–1186 (2005). [CrossRef] [PubMed]
17. L.H. Laiho, S. Plete, T.M. Hancewicz, P.D. Kaplan, and P.T.C. So, “Two-photon 3-D mapping of ex vivo human skin endogenous fluorescence species based on fluorescence emission spectra,” J. Biomed. Opt. 10, 0240161–10 (2005). [CrossRef] [PubMed]
18. B.M. Kim, J. Eichler, K.M. Reiser, A.M. Rubenchik, and L.B.D. Silva, “Collagen structure and nonlinear susceptibility: effects of heat, glycation, and enzymatic cleavage on second harmonic signal intensity,” Lasers Surg. Med. 27, 329–335 (2000). [CrossRef] [PubMed]
19. S.J. Lin, R.J. Wu, H.Y. Tan, W. Lo, W.C. Lin, T.H. Young, C.J. Hsu, J.S. Chen, S.H. Jee, and C.Y. Dong, “Evaluating cutaneous photoaging by use of multiphoton fluorescence and second-harmonic generation microscopy,” Opt. Lett. 17, 2275–2277 (2005). [CrossRef]
20. A.T. Yeh, M.J. Hammer-Wilson, D.C.V. Sickle, H.P. Benton, A. Zoumi, B.J. Tromberg, and G.M. Peavy, “Nonlinear optical microscopy of articular cartilage,” Osteoarthritis and Cartilage. 13, 345–352 (2005). [CrossRef] [PubMed]
21. H.S. Lee, Y. Liu, H.C. Chen, L.L. Chiou, G.T. Huang, W. Lo, and C.Y. Dong, “Optical biopsy of liver fibrosis by use of multiphoton microscopy,” Opt. Lett. 29, 2615–2616 (2004). [CrossRef]
22. J. Condeelis and J.E. Segall, “Intravital imaging of cell movement in tumours,” Nat.Rev. 3, 921–930 (2003). [CrossRef]
23. D.Y. Li, B. Brooke, E.C. Davis, R.P. Mecham, L.K. Sorensen, B.B. Boak, E. Eichwald, and M.T. Keating, “Elastin is an essential determinant of arterial morphogenesis,” Nature. 393, 276–280 (1998). [CrossRef] [PubMed]
25. A. Zoumi, X. Lu, G.S. Kassab, and B.J. Tromberg, “Imaging coronary artery microstructure using second-harmonic and two-photon fluorescence microscopy,” Biophys. J. 87, 2278–2286 (2004). [CrossRef]
29. P. Theer, M.T. Hasan, and W. Denk, “Two-photon imaging to a depth of 1000 microns in living brains by use of a Ti:Al203 regenerative amplifier,” Opt. Lett. 28, 1022–1024 (2003). [CrossRef] [PubMed]
30. H.J. van Staveren, C.J.M. Moes, J. van Marie, S.A. Prahl, and M.J.C. van Gemert, “Light scattering in intralipid-10% in the wavelength range of 400–1100 nm,” Appl .Opt. 30, 4057–4514 (1991).
31. A.K. Dunn, V.P. Wallace, M. Coleno, M.W. Berns, and B.J. Tromberg, “Influence of optical properties on two-photon fluorescence imaging in turbid samples,” Appl. Opt. 39, 1194–1201 (2000). [CrossRef]
32. P.T. Fwua, W. Loa, and C.Y. Dong, “Refractive index determination from multiphoton and confocal spherical aberration microscopy,” in Multiphoton Microscopy in the Biomedical Sciences V, A. Periasamy and P.T.C. So, eds., Proc. SPIE. 5700, 70–77 (2005). [CrossRef]
33. T. Collier, D. Arifler, A. Malpica, M. Follen, and R. Richards-Kortum, “Determination of epithelial tissue scattering coefficient using confocal microscopy,” IEEE J.Sele.Top. Qua. Elec. 9, 307–313 (2003). [CrossRef]
34. L. Moreaux, O. Saadre, and J. Mertz, “Membrane imaging by second-harmonic generation microscopy,” J. Opt. Soc. Am B, 17, 1685–1694(2000). [CrossRef]