Real-time visualization of melanin granules in normal human skin using combined multiphoton and reflectance confocal microscopy.

BACKGROUND: Recent advances in biomedical optics have enabled dermal and epidermal components to be visualized at subcellular resolution and assessed noninvasively. Multiphoton microscopy (MPM) and reflectance confocal microscopy (RCM) are noninvasive imaging modalities that have demonstrated promising results in imaging skin micromorphology, and which provide complementary information regarding skin components. This study assesses whether combined MPM/RCM can visualize intracellular and extracellular melanin granules in the epidermis and dermis of normal human skin. METHODS: We perform MPM and RCM imaging of in vivo and ex vivo skin in the infrared domain. The inherent three-dimensional optical sectioning capability of MPM/RCM is used to image high-contrast granular features across skin depths ranging from 50 to 90 µm. The optical images thus obtained were correlated with conventional histologic examination including melanin-specific staining of ex vivo specimens. RESULTS: MPM revealed highly fluorescent granular structures below the dermal-epidermal junction (DEJ) region. Histochemical staining also demonstrated melanin-containing granules that correlate well in size and location with the granular fluorescent structures observed in MPM. Furthermore, the MPM fluorescence excitation wavelength and RCM reflectance of cell culture-derived melanin were equivalent to those of the granules. CONCLUSION: This study suggests that MPM can noninvasively visualize and quantify subepidermal melanin in situ.

Label-free identification and characterization of murine hair follicle stem cells located in thin tissue sections with Raman micro-spectroscopy.

Stem cells offer tremendous opportunities for regenerative medicine. Over the past decade considerable research has taken place to identify and characterize the differentiation states of stem cells in culture. Raman micro-spectroscopy has emerged as an ideal technology since it is fast, nondestructive, and does not require potentially toxic dyes. Raman spectroscopy systems can also be incorporated into confocal microscope imaging systems allowing spectra to be obtained from below the tissue surface. Thus there is significant potential for monitoring stem cells in living tissue. Stem cells that reside in hair follicles are suitable for testing this possibility since they are close to the skin surface, and typically clustered around the bulge area. One of the first steps needed would be to obtain Raman micro-spectra from stem cells located in thin sections of tissue, and then see whether these spectra are clearly different from those of the surrounding differentiated cells. To facilitate this test, standard 5 µm thick sections of murine skin tissue were stained to identify the location of hair follicle stem cells and their progeny. Raman spectra were then obtained from adjacent cells in a subsequent unstained 10 µm thick section. The spectra revealed significant differences in peak intensities associated with nucleic acids, proteins, lipids and amino acids. Statistical analyses of the Raman micro-spectra identified stem cells with 98% sensitivity and 94% specificity, as compared with a CD34 immunostaining gold standard. Furthermore analyses of the spectral variance indicated differences in cellular dynamics between the two cell groups. This study shows that Raman micro-spectroscopy has a potential role in identifying adult follicle stem cells, laying the groundwork for future applications of hair follicle stem cells and other somatic stem cells in situ.

Differentiation of HaCaT cell and melanocyte from their malignant counterparts using micro-Raman spectroscopy guided by confocal imaging.

BACKGROUND: Skin cancer is the most common type of cancer in humans. Current techniques for identifying normal and neoplastic tissues are either destructive or not sensitive and specific enough. Raman spectroscopy and confocal imaging may obviate many limitations of existing methods by providing noninvasive, high-resolution, and real-time morphological and biochemical analysis of living tissues and cells. METHODS: We conducted micro-Raman spectroscopy studies on HaCaT cells, melanocytes (MC) and their malignant counterparts squamous cell carcinoma (SCC) and melanoma (MM) cells, respectively. Reflectance confocal imaging is used as guidance for the spectral measurements. RESULTS: Significant differences were found between the spectra of HaCaT cells and SCC cells, MC cells and MM cells, as well as all normal cells (HaCaT and MC) and all tumor cells (SCC and MM). Approximately 90% sensitivity and specificity was achieved for all the separations that we performed. CONCLUSION: Our results demonstrated the robust capability of confocal Raman spectroscopy in separating different cell lines. The acquired Raman spectra of major types of skin cells and their malignant counterparts will be useful for the interpretation of Raman spectra from in vivo skin. We believe it will eventually help diagnosis of skin cancer and other skin disease in clinical dermatology.

In vivo video rate multiphoton microscopy imaging of human skin.

We present a multiphoton microscopy instrument specially designed for in vivo dermatological use that is capable of imaging human skin at 27 frames per second with 256 pixels × 256 pixels resolution without the use of exogenous contrast agents. Imaging at fast frame rates is critical to reducing image blurring due to patient motion and to providing practically short clinical measurement times. Second harmonic generation and two-photon fluorescence images and videos acquired at optimized wavelengths are presented showing cellular and tissue structures from the skin surface down to the reticular dermis.

Near-infrared autofluorescence imaging of cutaneous melanins and human skin in vivo.

In recent years, near-infrared (NIR) autofluorescence imaging has been explored as a novel technique for tissue evaluation and diagnosis. We present an NIR fluorescence imaging system optimized for the dermatologic clinical setting, with particular utility for the direct characterization of cutaneous melanins in vivo. A 785-nm diode laser is coupled into a ring light guide to uniformly illuminate the skin. A bandpass filter is used to purify the laser light for fluorescence excitation, while a long-pass filter is used to block the main laser wavelength but pass the spontaneous components for NIR reflectance imaging. A computer-controlled filter holder is used to switch these two filters to select between reflectance and fluorescence imaging modes. Both the reflectance and fluorescence photons are collected by an NIR-sensitive charge-coupled device (CCD) camera to form the respective images. Preliminary results show that cutaneous melanin in pigmented skin disorders emits higher NIR autofluorescence than surrounding normal tissue. This confirmed our previous findings from NIR fluorescence spectroscopy study of cutaneous melanins and provides a new approach to directly image the distributions of cutaneous melanins in the skin. In-vivo NIR autofluorescence images may be useful for clinical evaluation and diagnosis of pigmented skin lesions, including melanoma.

Spectroscopic characterization and microscopic imaging of extracted and in situ cutaneous collagen and elastic tissue components under two-photon excitation.

BACKGROUND/PURPOSES: Understanding the two-photon excitation spectral characteristics and microscopic morphology of cutaneous collagen and elastic tissue components is important for applying multiphoton microscopy (MPM) in basic skin biology research and for clinical diagnosis. METHODS: We developed a system for two-photon excitation spectral measurements at various excitation wavelengths. The microscopic morphology was studied using a commercial multiphoton microscope. RESULTS: We obtained two-photon excitation fluorescence (TPEF) excitation-emission matrices (EEM), for the first time, of purified collagen and elastin samples, as well as in situ collagen and elastic fibers within excised human dermis. The EEM of the dermis was found to be similar to that of elastin. The excitation spectra for second harmonic generation (SHG) from purified collagen and excised dermis were also studied and were found to have similar spectral shapes. CONCLUSION: This study, using the EEM spectroscopic approach, confirmed a previous imaging study inference that in the dermis, TPEF predominantly originates from elastic fibers, while SHG originates solely from collagen fibers. The EEM data and SHG excitation spectra obtained in this study can be used to guide the selection of excitation wavelengths for MPM applications in basic skin biology research and for clinical diagnosis.

Integrated real-time Raman system for clinical in vivo skin analysis.

BACKGROUND: Raman spectroscopy is a non-invasive optical technique that can probe the molecular structure and conformation of biochemical constituents. The probability of Raman scattering is exceedingly low ( approximately 10(-10)), and consequently up to now the practical application of Raman spectroscopy to clinical medicine has been limited by either the weak spectral signal or by the long data acquisition times. Recent advances in Raman hardware and probe design have reduced spectral acquisition times, paving the way for clinical applications. METHODS: We present an integrated real-time Raman spectroscopy system for skin evaluation and characterization, which combines customized hardware features and software implementation. Real-time data acquisition and processing includes CCD dark-noise subtraction, wavelength calibration, spectral response calibration, intensity calibration, signal saturation detection, cosmic ray rejection, fluorescence background removal, and composition modeling. Real-time in vivo Raman measurement of volar forearm skin is presented to illustrate the methods and modeling. RESULTS: The system design implemented full-chip vertical hardware binning to improve the signal-to-noise ratio by 16-fold. The total time for a single in vivo measurement with analysis can be reduced to 100 ms with this implementation. Human skin was well modeled using the base Raman spectra. CONCLUSION: In vivo real-time Raman can be a very promising research and practical technique for skin evaluation.

Automated autofluorescence background subtraction algorithm for biomedical Raman spectroscopy.

A significant advantage of Raman spectroscopy as a noninvasive optical technique is its ability to detect subtle molecular or biochemical signatures within tissue. One of the major challenges for biomedical Raman spectroscopy is the removal of intrinsic autofluorescence background signals, which are usually a few orders of magnitude stronger than those arising from Raman scattering. A number of methods have been proposed for fluorescence background removal including excitation wavelength shifting, Fourier transformation, time gating, and simple or modified polynomial fitting. The single polynomial and the modified multi-polynomial fitting methods are relatively simple and effective, and thus are widely used in biological applications. However, their performance in real-time in vivo applications and low signal-to-noise ratio environments is sub-optimal. An improved automated algorithm for fluorescence removal has been developed based on modified multi-polynomial fitting, but with the addition of (1) a peak-removal procedure during the first iteration, and (2) a statistical method to account for signal noise effects. Experimental results demonstrate that this approach improves the automated rejection of the fluorescence background during real-time Raman spectroscopy and for in vivo measurements characterized by low signal-to-noise ratios.

Monte Carlo simulation of cutaneous reflectance and fluorescence measurements--the effect of melanin contents and localization.

Melanin content and distribution in skin were studied by examining a patient with white, brown and blue skin tones expressed on skin affected by vitiligo. Both diffuse reflectance and autofluorescence spectra of the three distinction skin sites were measured and compared. Monte Carlo simulations were then performed to help explain the measured spectral differences. The modeling is based on a six-layer skin optical model established from published skin optical parameters and by adding melanin content into different locations in the model skin. Both the reflectance and fluorescence spectra calculated by Monte Carlo (MC) simulation were approximately in agreement with experimental results. The study suggests that: (1) trichrome vitiligo skin may be an ideal in vivo model for studying the effect of skin melanin content and distribution on skin spectroscopy properties. (2) Based on the skin optical model and MC simulation, the content and distribution of melanin in skin, or other component of skin could be simulated and predicted. (3) Both reflectance and fluorescence spectra provided information about superficial skin structures but fluorescence spectra are capable of providing information from deeper cutaneous structures. (4) The research method, including the spectral ratio method, the method of adding and modifying the melanin content in skin optical models, and MC simulation could be applied in other non-invasive optical studies of the skin.

Cutaneous melanin exhibiting fluorescence emission under near-infrared light excitation.

Under ultraviolet and visible light excitation, melanin is essentially a nonfluorescent substance. This work reports our study on near-infrared (NIR) fluorescence properties of melanins, and explores potential applications of NIR fluorescence techniques for evaluating skin disorders involving melanin. The NIR fluorescence spectrum is obtained using a fiber optic NIR spectrometer under 785-nm laser excitation. In vitro measurements are performed on synthetic dihydroxyphenylalanine (DOPA) melanin, melanin extracted from Sepia ink sacs, human hair, animal fur, and bird feathers. Paired spectral comparisons of white and black skin appendages show that melanization of hair, fur, or feathers more than doubles the NIR fluorescence. In vivo NIR autofluorescence of normal dorsal and volar forearm skin of 52 volunteers is measured. Dorsal forearm skin, which is darker than volar skin, exhibits significantly greater NIR fluorescence. Patients with vitiligo (n=4), compound nevus (n=3), nevus of Ota (n=1), superficial spreading melanoma (n=3), and postinflammatory hyperpigmentation (n=1) are also evaluated. NIR fluorescence is greater within the lesion than the surrounding normal skin for all these conditions except vitiligo, where the converse was true. The observed melanin NIR fluorescence provides a new approach to in vitro and in vivo melanin detection and quantification that may be particularly useful for evaluating pigmented skin lesions.

Raman spectroscopy in combination with background near-infrared autofluorescence enhances the in vivo assessment of malignant tissues.

The diagnostic ability of optical spectroscopy techniques, including near-infrared (NIR) Raman spectroscopy, NIR autofluorescence spectroscopy and the composite Raman and NIR autofluorescence spectroscopy, for in vivo detection of malignant tumors was evaluated in this study. A murine tumor model, in which BALB/c mice were implanted with Meth-A fibrosarcoma cells into the subcutaneous region of the lower back, was used for this purpose. A rapid-acquisition dispersive-type NIR Raman system was employed for tissue Raman and NIR autofluorescence spectroscopic measurements at 785-nm laser excitation. High-quality in vivo NIR Raman spectra associated with an autofluorescence background from mouse skin and tumor tissue were acquired in 5 s. Multivariate statistical techniques, including principal component analysis (PCA) and linear discriminant analysis (LDA), were used to develop diagnostic algorithms for differentiating tumors from normal tissue based on their spectral features. Spectral classification of tumor tissue was tested using a leave-one-out, cross-validation method, and the receiver operating characteristic (ROC) curves were used to further evaluate the performance of diagnostic algorithms derived. Thirty-two in vivo Raman, NIR fluorescence and composite Raman and NIR fluorescence spectra were analyzed (16 normal, 16 tumors). Classification results obtained from cross-validation of the LDA model based on the three spectral data sets showed diagnostic sensitivities of 81.3%, 93.8% and 93.8%; specificities of 100%, 87.5% and 100%; and overall diagnostic accuracies of 90.6%, 90.6% and 96.9% respectively, for tumor identification. ROC curves showed that the most effective diagnostic algorithms were from the composite Raman and NIR autofluorescence techniques.

Counting moles automatically from back images.

Density of moles is a strong predictor of malignant melanoma, therefore, enumeration of moles is often an integral part of many studies that look at malignant melanoma. An automatic method of segmenting and counting moles would help standardize studies, compared with manual counting. We have developed an unsupervised algorithm for segmenting and counting moles from two-dimensional color images of the back torso region, as part of a study to evaluate the effectiveness of sunscreen. The method consists of a new variant of mean shift filtering that forms clusters in the image and removes noise, a region growing procedure to select candidates, and a rule-based classifier to identify moles. When this algorithm was compared to an assessment by an expert dermatologist, the algorithm showed a sensitivity rate of 91% and diagnostic accuracy of 90% on the test set, for moles larger than 1.5 mm in diameter.

Effect of formalin fixation on the near-infrared Raman spectroscopy of normal and cancerous human bronchial tissues.

Raman spectroscopy is a vibrational spectroscopic technique that can be used to probe molecular changes associated with tissue malignancy. In this report, the effect of formalin fixation on human bronchial tissues was studied by near-infrared (NIR) Raman spectroscopy to determine if the variations of Raman spectra caused by formalin fixation would affect the potential diagnostic ability for lung cancer detection. A rapid dispersive-type NIR Raman system with an excitation wavelength of 785 nm was used for this study. Bronchial tissue samples were obtained from six patients with known or suspected malignancies of the lung. Raman spectra of fresh normal and tumor tissue were compared with spectra of formalin-fixed normal and tumor tissue. Changes of the ratios of Raman intensities at 1445 to 1655 cm(-1) and 1302 to 1265 cm(-1) versus formalin fixing times varying from 2 to 24 h were also examined. The major tissue Raman peaks at 1265, 1302, 1445, and 1655 cm(-1) were found in both fresh and fixed bronchial tissues. However, bronchial tissue preserved in formalin showed a progressive decrease in overall intensities of these Raman peaks. Raman contaminations due to formalin were also found in the 980-1100, and 1480-1650 cm(-1) ranges with notable formalin peaks (1041 and 1492 cm(-1)) appearing in the fixed normal and tumor tissues. The results showed that NIR Raman spectra of human bronchial tissues were significantly affected by formalin fixing and tissue hydration. Diagnostic markers at the 980-1100, and 1500-1650 cm(-1) regions derived from fixed tissues do not appear to be applicable for in vivo lung cancer detection. To yield valid Raman diagnostic information for in vivo applications, fresh tissue should be used. If only fixed tissue is available, thorough rinsing of specimens in phosphate-buffered saline (PBS) before spectral measurements may help reduce the formalin fixation artifacts on tissue Raman spectra.

Laser-induced autofluorescence microscopy of normal and tumor human colonic tissue.

Laser-induced autofluorescence (LIAF) spectroscopy has been found to be a promising tool for early cancer diagnosis in various organs, but the reasons responsible for the spectral differences between normal and diseased tissue are still not well understood. In this study, a microspectrophotometer (MSP) system was used to identify the microscopic origins of tissue autofluorescence in the colon under the excitation of a helium-cadmium laser at 442 nm. Colonic tissue samples (normal: n=8, adenocarcinoma: n=10) were obtained from 12 patients with known or suspected malignancies of the colon. The intrinsic fluorescence spectra and images of fresh tissue sections prepared from normal and tumor colonic tissue were measured by the MSP system. Three distinct tissue layers of the colon were found for fluorescence, the mucosa, the submucosa and the muscularis propria, with submucosa being the most fluorescent. Differences in the spectral shape and intensity of the intrinsic fluorescence originating from different colonic layers indicate that fundamentally different fluorophores may be present in the respective tissue layers. There was no significant difference in the intrinsic fluorescence features of the submucosa between normal and tumor colonic tissue, but the fluorescence intensity of the submucosa in tumor tissue was significantly reduced due to the infiltration of tumor cells into the submucosa. The intrinsic fluorescence spectrum peaking at about 520 nm for tumor stroma appeared more evident than that of normal lamina propria. Limited areas of the lamina propria layer in some adenocarcinoma colon exhibited an emission band at about 635 nm, which was attributed to endogenous porphyrins in tumor. Autofluorescence microscopy revealed that differences in the clinically measured autofluorescence spectra between normal and tumor tissue were mainly due to thickening of the tumor mucosa resulting in a reduced submucosa fluorescence contribution, as well as the increased hemoglobin absorption in tumor tissue. Therefore, investigation of the microscopic origins of tissue autofluorescence and images can provide new insights into morphological structures and biochemical components of tissues, which are vital to improve the implementation of the LIAF technique for non-invasive in vivo tissue diagnostics.

Raman spectroscopy of in vivo cutaneous melanin.

We successfully acquire the in vivo Raman spectrum of melanin from human skin using a rapid near-infrared (NIR) Raman spectrometer. The Raman signals of in vivo cutaneous melanin are similar to those observed from natural and synthetic eumelanins. The melanin Raman spectrum is dominated by two intense and broad peaks at about 1580 and 1380 cm(-1), which can be interpreted as originating from the in-plane stretching of the aromatic rings and the linear stretching of the C-C bonds within the rings, along with some contributions from the C-H vibrations in the methyl and methylene groups. Variations in the peak frequencies and bandwidths of these two Raman signals due to differing biological environments have been observed in melanin from different sources. The ability to acquire these unique in vivo melanin signals suggests that Raman spectroscopy may be a useful clinical method for noninvasive in situ analysis and diagnosis of the skin.

Monitoring photoproduct formation and photobleaching by fluorescence spectroscopy has the potential to improve PDT dosimetry with a verteporfin-like photosensitizer.

In current clinical practice, photodynamic therapy (PDT) is carried out with prescribed drug doses and light doses as well as fixed drug-light intervals and illumination fluence rates. This approach can result in undesirable treatment outcomes of either overtreatment or undertreatment because of biological variations between different lesions and patients. In this study, we explore the possibility of improving PDT dosimetry by monitoring drug photobleaching and photoproduct formation. The study involved 60 mice receiving the same drug dose of a novel verteporfin-like photosensitizer, QLT0074, at 0.3 mg/kg body weight, followed by different light doses of 5, 10, 20, 30, 40 or 50 J/cm2 at 686 nm and a fluence rate of 70 mW/cm2. Photobleaching and photoproduct formation were measured simultaneously, using fluorescence spectroscopy. A ratio technique for data processing was introduced to reliably detect the photoproduct formed by PDT on mouse skin in vivo. The study showed that the QLT0074 photoproduct is stable and can be reliably quantified. Three new parameters, photoproduct score (PPS), photobleaching score (PBS) and percentage photobleaching score (PBS%), were introduced and tested together with the conventional dosimetry parameter, light dose, for performance on predicting PDT-induced outcome, skin necrosis. The statistical analysis of experimental results was performed with an ordinal logistic regression model. We demonstrated that both PPS and PBS improved the prediction of skin necrosis dramatically compared to light dose. PPS was identified as the best single parameter for predicting the PDT outcome.

Irregularity index: a new border irregularity measure for cutaneous melanocytic lesions.

One of the important clinical features that differentiates benign melanocytic nevi from malignant melanomas is the irregularity of the lesion border. There are two types of border irregularities: texture irregularities, the small variations along the border, and structure irregularities, the global indentations and protrusions. Texture irregularities are subject to noise, whereas structure irregularities may suggest excessive cell growth or regression of a melanoma. We have designed a new algorithm for measuring the structure irregularities in the border. Our algorithm first locates all the local and global indentations and protrusions and organizes them in a hierarchical structure. Then an area-based index, called the irregularity index, is computed for each indentation and protrusion along the border. From the individual irregularity indices, two important new measures, the most significant irregularity index and the overall irregularity index are derived. These two new indices provide a measure of the degree of irregularity along the lesion border. A double-blinded test was performed to examine the effectiveness of these two new indices. Fourteen experienced dermatologists were asked to evaluate the borders of 40 pigmented lesions. The clinical evaluation result was then compared with the two new indices and other published shape measurements. The user study showed that both of the new indices vastly outperformed the other shape descriptors. Moreover, our algorithm captured the knowledge of expert dermatologists in analysing malignancy of a lesion based on its shape alone, indicating that the new measures may be useful for diagnosing melanomas.

Durable rough skin phantoms for optical modeling.

Skin phantoms are often used to study and model light propagation. However, existing skin phantoms overlook the important effect of surface roughness on light propagation patterns. This paper reports the construction of durable phantoms with controllable surface roughness and bulk optical properties. With silica microspheres as the scattering particles, we theoretically model the scatterer density required to achieve the desired phantom optical properties before fabrication. The surface roughness and the attenuation coefficients of the constructed phantoms were validated using optical profilometry and ballistic spatial filter photometry. These rough skin phantoms were originally developed for laser speckle studies, but could also be used for studying optical phenomena where light experiences surface and bulk scattering at the same time.

Imaging directed photothermolysis through two-photon absorption demonstrated on mouse skin - a potential novel tool for highly targeted skin treatment.

One-photon absorption based traditional laser treatment may not necessarily be selective at the microscopic level, thus could result in un-intended tissue damage. Our objective is to test whether two-photon absorption (TPA) could provide highly targeted tissue alteration of specific region of interest without damaging surrounding tissues. TPA based laser treatments (785 nm, 140 fs pulse width, 90 MHz) were performed on ex vivo mouse skin using different average power levels and irradiation times. Reflectance confocal microscopy (RCM) and combined second-harmonic-generation (SHG) and two-photon fluorescence (TPF) imaging channels were used to image before, during, and after each laser treatment. The skin was fixed, sectioned and H & E stained after each experiment for histological assessment of tissue alterations and for comparison with the non-invasive imaging assessments. Localized destruction of dermal fibers was observed without discernible epidermal damage on both RCM and SHG + TPF images for all the experiments. RCM and SHG + TPF images correlated well with conventional histological examination. This work demonstrated that TPA-based light treatment provides highly localized intradermal tissue alteration. With further studies on optimizing laser treatment parameters, this two-photon absorption photothermolysis method could potentially be applied in clinical dermatology.

A method for accurate in vivo micro-Raman spectroscopic measurements under guidance of advanced microscopy imaging.

The movement from the subjects during in vivo confocal Raman spectral measurements could change the measurement volume, leading to non-specific signals and inaccurate interpretation of the acquired spectrum. Here we introduce a generally applicable method that includes (1) developing a multimodal system to achieve real-time monitoring of every spectral measurement with reflectance confocal microscopy (RCM) and multiphoton microscopy (MPM) imaging; (2) performing region-of-interest measurement by scanning an area of the tissue during spectral acquisition. The developed method has been validated by measuring different micro-structures of in vivo human skin. Our results demonstrated great consistency between RCM images and confocal Raman spectra. The superior quality of the images and spectra allows us to derive blood flow velocity and blood glucose level. We believe this method is valuable for realizing accurate microscopic spectral measurement and have great potential to be adapted into clinic to achieve non-invasive measurement of important biological parameters.