Base line spectral excitation and emission scans were defined for the oral mucosa in a population of 61 controls, 16 oral tongue cancer patients and 2 patients with tongue leukoplakia. A xenon-based fluorescence spectrophotometer (Mediscience Corp.) with a fiberoptic probe (Mediscience Corp.) was used to collect excitation and emission spectra. Two excitation scans ((lambda) Ex 200-360 nm, (lambda) Em 380 nm; (lambda) Ex 240-430 nm, (lambda) Em 450 nm) and two emission scans ((lambda) Ex 300 nm, (lambda) Em 320-580 nm; (lambda) Ex 340 nm, (lambda) Em 360-660 nm) were used to analyze the buccal mucosa (BM), hard palate (HP), floor of mouth (FOM) and dorsal tongue (DT) of 61 control individuals. In 41 controls the lateral tongue site (LT) was added. The same set of scans was performed on tumor lesions and contralateral normal tissues of 16 patients with lateral tongue tumors and on two individuals with leukoplakia of the tongue. Ratios of points on the individual scans were used to quantitate data. The excitation scan ((lambda) Ex 200-360 nm, (lambda) Em 380 nm) and the emission scan ((lambda) Ex 300 nm, (lambda) Em 320-580 nm) were able to statistically discriminate the HP and DT from the BM and FOM. The ratios of intensities of neoplastic mucosa and contralateral sites were significantly different with the excitation scans ((lambda) Ex 200-360 nm, (lambda) Em 380 nm, p < 0.001) and ((lambda) Ex 240-430 nm, (lambda) Em 450 nm, p < 0.01) and with the emission scan ((lambda) Ex 300 nm, (lambda) Em 320-580 nm, p < 0.001). Discrimination was significant with the emission scan ((lambda) Ex 340 nm, (lambda) Em 360- 660 nm, p < 0.07). Innate tissue fluorescence has potential as a monitor of cancer patients and populations at risk for head and neck cancer.

Laser-induced fluorescence (LIF) was used for direct in-vivo cancer diagnosis of the esophagus without requiring biopsy. The methodology was applied to differentiate normal and malignant tumors of the esophagus. Endogenous fluorescence of normal and malignant tissues were measured directly using a fiberoptic probe inserted through an endoscope. The measurements were performed in vivo during routine endoscopy. Detection of the fluorescence signal from the tissue was performed using laser excitation. The results of this LIF approach were compared with histopathology results of the biopsy samples and indicated excellent agreement in the classification of normal and malignant tumors for the samples investigated. The LIF procedure could lead to the development of a rapid and cost-effective technique for cancer diagnosis.

Photodynamic therapy (PDT) and photodynamic diagnosis (PDD) add support to efficient treatment modalities of superficial and early stage cancer. Recently 5-aminolevulinic acid (5-ALA), a precursor of hemoglobin in the hem biosynthetic pathway, was used to stimulate endogenous porphyrin production. The time dependency of 5-ALA induced porphyrin fluorescence has been investigated on several normal tissues as well as on a tumor in an in-vivo tumor model (human gastrointestinal adenocarcinoma Grade II, UICC IIa). 5-ALA has been administered intravenously at a concentration of 50 mg/(kg bw). With respect to a certain time schedule the animals were sacrificed and 12 different organs as well as the tumor were removed. Using laser-induced fluorescence techniques the emission spectra in the range of (lambda) equals (550-750) nm were detected from the tissues after excitation with light of the wavelength (lambda) equals (411 +/- 4) nm. For quantitative evaluation the integral fluorescence intensity at (lambda) equals (635 +/- 2) nm of the porphyrin specific spectra has been determined. All tissues showed porphyrin fluorescence, while brightest fluorescence has been detected from the tumor. With respect to the other tissues the relative tumor selectivity showed a maximum ratio at 406 h post injection. The kinetics of the porphyrin fluorescence intensity of the organs follow different time dependencies. Simple mathematical pharmacokinetic models are developed and discussed.

Laser-induced fluorescence was used for the monitoring of the (delta) - amino levulinic acid (ALA)-induce protoporphyrin IX (PpIX) build-up in nonmelanoma malignant tumors of the skin and some cancers in the head and neck region. An optical-fiber based point monitoring system was utilized in the recording of fluorescence spectral at different time intervals after the administration of ALA. In the cases of skin tumors ALA was normally applied topically to the area. Only in one patient with an aggressive skin tumor ALA was administered intravenously. For the PpIX induction in the head and neck tumors the ALA was given orally. An example of a tumor fluorescence image is also presented.

In vitro cell model systems have been used to study the mechanisms of intrinsic cellular and tissue fluorescence as a potential biomarker for cancer. Phenotypic characteristics of cancer that are different from normal tissue include changes in histoarchitecture, proliferation rates and differentiation. a nitrosmethlybenzylamine (NMBA)/rat esophageal carcinogenesis model (NMBA), a transforming growth factor beta (TGF- (beta) )/normal epithelial cell model, and a retinoic acid (RA)/multicellular tumor spheroid model (RAMTS) were used to assess fluorescence changes associated respectively with changes in histoarchitecture, proliferation rates and differentiation. A xenon based fluorescence spectrophotometer (Mediscience Corp.) was used to collect excitation and emission spectra. Two excitation scans ((lambda) Ex 200-360 nm, (lambda) Em 380 nm; (lambda) Ex 240-430 nm, (lambda) Em 450 nm) and two emission scans ((lambda) Ex 300 nm, (lambda) Em 320-580 nm; (lambda) Ex 340 nm, (lambda) Em 360-660 nm) were used to analyze the three model systems. Using the NMBA model. Differences were seen in the excitation scan ((lambda) Ex 200-360 nm, (lambda) Em 380 nm) and the emission scan ((lambda) Ex 340 nm, (lambda) Em 360-660 nm) when normal rat esophageal tissue was compared to hyperplastic and tumor tissue. In the (TGF-(beta) ) model, differences were seen in the excitation scan ((lambda) Ex 240-430 nm, (lambda) Em 450 nm) when comparing proliferation slowed (TGF-(beta) treated) epithelial cells to their untreated controls. In the RAMTS model, differences were seen with all four scans when RA treated multicellular tumor spheroids (nondifferentiating) were compared to untreated control cells (differentiating). The data indicate that fluorescence changes seen in these model systems may relate to changes in histoarchitecture, proliferation rates and differentiation. Their relationship to in vivo fluorescence changes seen in cancer patients remains to be elucidated.

A miniature spectrometer system has been constructed for both reflectance and autofluorescence spectral measurements of skin. The system is based on PC plug-in spectrometer, therefore, it is miniature and easy to operate. The spectrometer has been used clinically to collect spectral data from various skin lesions including skin cancer. To date, 48 patients with a total of 71 diseased skin sites have been measured. Analysis of these preliminary data suggests that unique spectral characteristics exist for certain types of skin lesions, i.e. seborrheic keratosis, psoriasis, etc.. These spectral characteristics will help the differential diagnosis in Dermatology practice. In conjunction with the spectral point measurements, we are building and testing a multispectral imaging system to measure the spatial distribution of skin reflectance and autofluorescence. Preliminary results indicate that a cutaneous squamous cell carcinoma has a weak autofluorescence signal at the edge of the lesion, but a higher autofluorescence signal in the central area.

The origin of fluorescent or phosphorescent signals re-emitted from highly scattering media (such as tissues) is investigated using a simple probability analysis developed earlier by Seveck et al. and adapted for incorporation of excited state probe kinetics. Results show that the lifetime of a uniformly distributed phosphorescent or fluorescent optical probe will profoundly affect the volume interrogated by a noninvasive, reflectance measurements. When the lifetime is greater than photon migration times, the origin of the re-emitted signal is confined closest to the surface. Our computations suggest that noninvasive measurements of tissue oxygen concentration may not necessarily interrogate deep tissues when systemically administered phosphorescent dyes are used. Instead, our results point to the development of short- lived probes and the successful deconvolution of nanosecond photon migration times for sensing subsurface conditions.

The in vivo laser induced fluorescence experiments have shown that the intrinsic LIF spectral shape of bronchial dysplasia and carcinoma in situ does not differ significantly from that of normal bronchial tissue. However, the intensity of LIF from diseased tissue is significantly reduced. Comparing the in vivo measurement results and the Monte Carlo modeling based on measured optical properties, we have hypothesized that the factors which cause the fluorescence reduction in diseased tissue may be (i) thicker abnormal epithelium, (ii) slightly higher blood content, and (iii) reduction in the density of the fluorescence source density of submucosa. Based on the above hypothesis, the Monte Carlo simulation code was modified to allow one to model spectrally distorted (when compared to the intrinsic fluorescence spectra) fluorescence collected by an imaging system. The fluorescence images excited by He-Cd laser (442 nm) at three typical wavelengths (500 nm, 577 nm and 630 nm) were analyzed by using Monte Carlo simulations. The results indicated that even if there is not significant difference in spectral shape of intrinsic fluorescence between normal and abnormal bronchial tissue, the ratio of red (630 nm) to green (500 nm) fluorescence image from the modeled normal site is significantly higher than that from modeled abnormal site. And abnormal tissue can be identified by the ratio imaging technique. The major factor causing the increased red/green ratio at abnormal tissue sites is the different tissue optics between normal and abnormal tissue. In the in vivo study, five bronchial carcinoma in situ sites and 17 moderate/sever dysplasia lesions were examined by a ratio fluorescence imaging system through a bronchoscope. The red/green ratios of carcinoma in situ sites and dysplasia sites were found to be 1.85 +/- 0.2 and 1.7 +/- 0.2 higher than that of normal sites. This was consistent with the simulation results.

Bladder tissue autofluorescence spectra are obtained in vivo at two excitation wavelengths (334 nm and 365 nm) with a cystoscopic fiber- optic device based on a small mercury arc lamp. Upon 365 nm excitation, both normal and cancerous bladder tissue have nearly identical fluorescence spectra, characterized by a broad peak at 455 nm. However, the fluorescence yield from malignant tissue is approximately a factor 3 lower compared to normal tissue. A similar decrease in fluorescence yield is observed upon 334 nm excitation. More importantly, at this excitation wavelength, the spectra from normal and malignant tissue have different lineshape. Normal tissue shows two distinct fluorescence peaks (at 385 nm and 455 nm), while malignant tissue only shows the 455 nm peak. Based on these insights, we have developed a simple spectroscopic algorithm, to differentiate normal from malignant bladder tissue with our device. The main underlying biophysics will be addressed. The integration of the diagnostic method with a reliable therapeutic technique for tumor cell destruction, may open the way for cost- effective preventive care of high-risk patients.

The mechanisms of optical contrast in biological tissues and the limits of contrast are discussed. Some cancers develop as cellular masses which displace normal fibrous tissue or stroma, yielding a net drop in scattering properties. Some cancer may be more fibrous than normal tissues yielding a net increase in scattering. Cancers sometimes develop increased vascularity which yields a net increase in absorption. Another measure is fluorescence, either autofluorescence or dye-enhanced fluorescence. This paper considers the limits of resolution when optically characterizing a small spherical object within a turbid tissue. The ability to discriminate a small object with a strongly increased absorption vs a larger object with a slight increase in absorption is tested, using frequency-domain light transport with 3-GHz modulation of the light source. The results suggest that an object about 1/10 the size of the tissue within which it is embedded can be distinguished by a pair of measurements: the phase difference (Delta) P and the amplitude ratio A/A₀.

Reflectance spectrophotometry allows an objective evaluation of surface colors and it has been proposed as a tool to discriminate cutaneous melanoma from other pigmented cutaneous lesions. In order to improve reflectance spectral analysis of moles, a novel spectrophotometric system has been developed, based on the use of a CCD camera provided with a set of interference filters. For each filter the system acquires an image of the lesion. The images are digitized and stored in a PC for off-line data handling. Reflectance images from visible to near IR were acquired of 51 cutaneous pigmented lesions including melanoma and dysplastic, compound and junctional nevus. Four reflectance standards were used to convert the image raw numbers into absolute reflectance values. From each spectral image, corresponding to a selected wavelength, four parameters were derived to quantify color, pigment distribution, dimension and boundary of each lesion. A statistical analysis performed on the selected parameters shows that the mean effective reflectance, pigment distribution and lesion area are significantly different when melanoma is compared to the other lesions. This result suggests that telespectrophotometry could be used as an aid in performing clinical diagnosis of melanoma.

Time-gated transillumination techniques for detecting and resolving regions of abnormal absorption and/or scattering in a slab geometry are based on the ability to detect transmitted photons that are nearly ballistic, i.e., that arrive at relatively short times after initial injection at a detector. However, at short gating times, the detectable intensity quickly falls below a practical detection threshold with increasing slab thickness. A quantitative theory of this effect is easily developed within the framework of random walk theory. Such a theoretical development has been shown to yield quite good approximations for properties of photon transport in a homogeneous multiply-scattering medium. Mathematical solutions of the random walk model are used to suggest experimentally measurable parameters with potential applicability tot he detection of abnormalities when their optical coefficients differ form those of the background.

The prospects for time-resolved optical mammography rests on the ability to detect adenocarcinoma within the breast with sufficient resolution and specificity to compete with X-ray mammography. We characterized the optical properties of an unusually large (6 cm diameter) fresh adenocarcinoma and normal breast tissue (determined by histology to be predominantly adipose tissue) obtained from a patient undergoing mastectomy. Large specimens (5 mm thick and 3 cm wide) allowed the determination of absorption and scattering coefficients and their spatial heterogeneity as probed with a 1 mm diameter laser beam at 633 nm and 800 nm utilizing total reflectance and transmittance measure with integrating spheres. The difference between scattering coefficients of the malignant tumor and those of normal (principally adipose) breast tissue at 633 nm was much greater than the heterogeneity within each sample. This scattering difference is the principal source of contrast, particularly in time-resolved images. However, the high scattering coefficient of normal breast tissue at 633 nm limits the practicality of time-resolved mammography of a human breast compressed to 5 cm. Although the scattering coefficient of the normal breast tissue decreases at 800 nm, the differences between the optical properties of normal and abnormal breast tissue also are reduced. We used these empirical results in theoretical expressions obtained from random walk theory to quantify the expected resolution, contrast, and the detected intensity of 3, 6, and 9 mm tumors within otherwise homogeneous human breasts as a function of the gating-time of time-resolved optical mammography.

Fluorescence spectra of dermal lesions on 48 rats have been investigated using excitation wavelengths of 275 nm, 300 nm and 340 nm. Emission at 340 nm and 460 nm were measured in both forequarter and hindquarter lesions. Unlike 460 nm emission, intensity at 340 nm increased with time and then saturated. Control studies on intact skin and lesions in dead rats failed to demonstrate any time dependent changes. It appears that the 340 nm intensity changes is due to changes in the tryptophan level, and may reflect a factor in the wound healing process.

Liposomes are known to be uptaken by the liver cells after intraveinous injection. Only few techniques are available to follow this process in vivo like nuclear magnetic resonance spectroscopy or scintigraphy. Intracellular pathway and liposomes localization in the different liver cells require sacrifice of the animals, cells separation and electronic microscopy, then little is known about liposomes kinetic uptake by the acidic intracellular compartments in vivo. We propose in this study a new method to follow liposomes uptake in the liver in vivo using a fluorescent pH sensitive probe 5,6-carboxyfluorescein and two different composition of liposomes: phospholipids DSPC/Chol and DMPC in order to evaluate the influence of the formulation on the release characteristics of liposomes in the lysosomes. We have already demonstrated the ability of the fluorescence spectroscopy and imaging using a pH dependent probe to monitor pH in living tissues. As pH of lysosomes is very low, the kinetic liposomes uptake in this intracellular acidic compartment is followed by monitoring the pH of the whole liver in vivo and ex vivo. Carboxyfluorescein is used at high concentration (100 mM) in order to quench its fluorescence. Liposomes are injected to Wistar rats into the penil vein. After laparotomy, fluorescence spectra and images are recorded during two hours. Results show a clear relationship between formulation of liposomes and stability in the acidic compartments of hepatic cells. After sacrifice and flush with cold saline solution, pH of the liver ex vivo is found to be 5.0-5.5. Data show a rapid clearance of release dye and an uptake of liposomes by the liver cells and, as liposomes penetrate in the acidic compartment, dye is released from liposomes and is delivered in lysosomes leading to the decrease of the pH.

Optical properties of frozen-ground and intact soft tissues obtained from spectrophotometer measurements of reflectance and transmittance were compared. The tissues used in these experiments were calf aorta, rat jejunum, and rabbit sciatic nerve. Each tissue specimen was frozen in liquid nitrogen and then ground with pestle and mortar into a fine powder. A tissue paste formed once the powder returned to room temperature. The tissue paste was then sandwiched between glass slides for spectrophotometric measurements. For comparison, the optical properties of the intact tissues were also measured. Total transmission and diffuse reflection were obtained on a Varian Cary 5 Spectrophotometer (400 nm - 850 nm). Absorption and reduced scattering coefficients of the tissues were determined with the Inverse Adding Doubling Method. Our results suggested that optical properties of soft tissue can be estimated from that of the ground tissue in the visible range.

We have developed a new intracavity laser technique that uses living or fixed cells at integral components of a laser. The cells are placed on an AlGaAs/GaAs surface-emitting semiconductor wafer and covered with a glass dielectric mirror to form a laser resonator. In this arrangement, the cells serve as optical waveguides (or lens elements) to confine (or focus) light generated in the resonator by the semiconductor. Because of the high transparency, the cells aid the lasing process to generate laser light. This ultra sensitive laser provides a novel imaging/spectroscopic technique for histologic examination which we demonstrate with normal and sickled human red blood cells. Extremely high contrast microscopic images of the cells are observed near 830-850 nm. These images correspond to electromagnetic modes of cell structures and are sensitive to shape of the cell. Using a high resolution spectrometer, we resolve the light emitted from these images into very narrow spectral peaks associated with the lasing modes. Analysis of the spectra reveals that the distribution of peaks is quite different for normal and sickled red blood cells. This technique, in a more developed form, may be useful for the rapid analysis of these and other kinds of normal and abnormal cells.

We have been interested in the role of Human Papillomavirus (HPV) in cervical cancer and its diagnosis; to that end we have been developing microscopic imaging and fluorescent in situ hybridization (FISH) techniques to genotype and quantitate the amount of HPV present at a single cell level in cervical PAP smears. However, we have found that low levels of HPV DNA are difficult to detect accurately because theoretically obtainable sensitivity is never achieved due to nonspecific autofluorescence, fixative induced fluorescence of cells and tissues, and autofluorescence of the optical components in the microscopic system. In addition, the absorption stains used for PAP smears are intensely autofluorescent. Autofluorescence is a rapidly decaying process with lifetimes in the range of 1-100 nsec, whereas phosphorescence and delayed fluorescence have lifetimes in the range of 1 microsecond(s) ec-10 msec. The ability to discriminate between specific fluorescence and autofluorescence in the time-domain has improved the sensitivity of diagnostic test such that they perform comparably to, or even more sensitive than radioisotopic assays. We have developed a novel time-resolved fluorescence microscope to improve the sensitivity of detection of specific molecules of interest in slide based specimens. This time-resolved fluorescence microscope is based on our recently developed fluorescence lifetime imaging microscopy (FILM) in conjunction with the use of long lifetime fluorescent labels. By using fluorescence in situ hybridization and the long lifetime probe (europium), we have demonstrated the utility of this technique for detection of HPV DNA in cervicovaginal cells. Our results indicate that the use of time-resolved fluorescence microscopy and long lifetime probes increases the sensitivity of detection by removing autofluorescence and will thus lead to improved early diagnosis of cervical cancer. Since the highly sensitive detection of DNA in clinical samples using fluorescence in situ hybridization image is useful for the diagnosis of many other type of diseases, the system we have developed should find numerous applications for the diagnosis of disease states.

Fluorescence spectra of tryptophan and bacteria were measured at different concentrations using a Mediscience CD-Scan unit. The emission spectra of tryptophan were obtained using an excitation wavelength at 280 nm. The excitation spectra were obtained at the emission of 340 nm. The minimum detectable concentration of tryptophan was found to be 10^-8 M. The emission spectra for bacteria were probed at 340 nm. The minimum detectable number of bacteria in a beam of the excitation light was determined to be about 30. Assuming that the emission band at 340 nm of bacteria comes from tryptophan, the number of tryptophan per bacterium was estimated to be 10⁸. This approach to determine the number is almost consistent with that obtained using a weight method.

High Resolution digital acquisition allows a great deal of flexibility in the types of questions that can be directed to microscopic samples. To eliminate subjective bias and provide quantitative results we have approached microscopy with an automated digital format. This mode can return quantitative data at high resolution over large fields. The digital format makes accessible data including [data segmentation]: multispectral colocalization, seeding and connectivity, particle size and shape distribution and population analysis. We have begun a program to investigate this approach using the confocal microscope. Scanning larger fields-of-view at lower spatial resolutions (e.g., low magnification objective) defines large maps that allow alignment of high spatial resolution (diffraction limited) sampling. The [objective] selection of the field-of-view with low spatial resolution reduces the subjective nature of the selection of a 'typical staining pattern'. High resolution digital scanning in three dimensions contribute both to the 'objective' nature of the analysis and allow for quantitation of characteristics not historically available/accessible. The complex carbohydrate heparin is implicated in tumor growth and wound healing by affecting angiogenesis, cell proliferation and motility. The internal localization of heparin within vascular cells appears to be a good predictor of the sensitivity of those cells to the action of heparin. Cells resistant to the antiproliferative action of heparin are able to sequester the heparin in large vacuoles whereas those cells sensitive to the carbohydrate do not exhibit these structures. We have applied our approach to QUANTITATIVE DIGITAL MICROSCOPY to the analysis of intracellular heparin distribution.

Possibilities to retrieve the erythrocyte size distribution and other parameters of the whole blood from the angular distribution of multiple scattered light are discussed. Experimental measurements of the small- angular distributions of laser radiation scattered in the blood layers of various optical depths and their interpretation are presented. The narrow peak of the coherent backward scattering of laser radiation for the whole blood is first observed. Its application to retrieval of blood parameters is discussed.

Three-dimensional confocal visualization of living human skin is a new development in the noninvasive imaging of normal and pathological tissue. I have investigated the autofluorescence of in vivo human skin with a laser scanning confocal microscope. An argon ion laser (488 nm) was used for excitation of the natural fluorescence of skin and a 515 nm cut off filter was used to separate the fluorescence from the excitation light. I found that normal skin has a very high autofluorescence. The laser scanning confocal microscope was used to obtain a stack of serial sections through the skin. A stack of optical sections through the hair follicle was reconstructed as well as the three-dimensional reconstruction of the pores of sweat glands. The ability to obtain two and three-dimensional visualizations of in vivo human skin may provide a new tool for noninvasive diagnostics in dermatology.

The mechanisms responsible for contrast in reflection mode imaging of turbid tissues at 1300 nm with an optical coherence microscope are addressed. A basic model is motivated and presented in which tissue backscatter is assumed to originate from substantially subwavelength scale refractive index fluctuations, while beam attenuation and decorrelation are attributed to structures larger than a wavelength. The sources of noise in OCM images are considered, particularly speckle, the behavior of which is calculated using the assumed properties of tissue. The speckle behavior is then used to estimate the contrast-detail performance of an OCM. The effects of multiple scattering on resolution and contrast are calculated numerically using a hybrid Monte Carlo/analytical model.

Image quality of a microscopic bar chart hidden behind a slab of highly scattering media was significantly improved by incorporating a spatial filter at the back Fourier (focal) plane of the objective lens of a conventional microscope. The image quality was further improved by combining a spatial filtering with time-resolved detection.

The Los Alamos National Laboratory has continued the development of the Optical Biopsy System (OBS) for noninvasive, real-time in situ diagnosis of tissue pathologies. In proceedings of earlier SPIE conferences we reported on clinical measurements in the bladder, and we report here on recent results of clinical tests in the gastrointestinal tract. With the OBS, tissue pathologies are detected/diagnosed using spectral measurement of the elastic optical transport properties (scattering and absorption) of the tissue over a wide range of wavelengths. The use of elastic scattering as the key to optical tissue diagnostics in the OBS is based on the fact that many tissue pathologies, including a majority of cancer forms, exhibit significant architectural changes at the cellular and subcellular level. Since the cellular components that cause elastic scattering have dimensions typically on the order of visible to near-IR wavelengths, the elastic (Mie) scattering properties will be wavelength dependent. Thus, morphology and size changes can be expected to cause significant changes in an optical signature that is derived from the wavelength-dependence of elastic scattering. Additionally, the optical geometry of the OBS beneficially enhances its sensitivity for measuring absorption bands. The OBS employs a small fiber optic probe that is amenable to use with any endoscope or catheter, or to direct surface examination, as well as interstitial needle insertion. Data acquisition/display time is < 1 second.

Changes in oxidized (HbO), reduced (Hbr), and total hemoglobin (Hbt) contents were monitored by near infrared spectroscopy in human forearm skeletal muscle during oscillatory variations of the effective venous pressure. Laser diode pulses (wavelength, 775, 810, 865, 904; duration, 60 ns) were directed to the muscle by means of an optic fiber bundle and detected with a sensing fiber positioned at 1.5 cm from the emitting bundle. Sinusoidal pressure waves (frequency, 1 and 2 cycles/min; amplitude, 10-15 mm Hg) generated by a piston pump were transmitted to a sphygmomanometer cuff placed on the arm, the mean cuff pressure (Pc) being 20, 40 and 60 mm Hg. Variations of HbO, Hbr and Hbt were computed from the optical signals and processed by Fourier analysis to characterize their amplitude and phase relative to the cuff mean pressure oscillations (Posc). Oscillations of HbO, Hbr, Hbt were observed for all mean cuff pressures, the amplitude of the variations being decreased with increasing Pc. For Pc equals 20 mm Hg, the oscillations of HbO and Hbt were nearly in phase with Posc whereas the oscillation of Hbr were out of phase with HbO and Posc. Increasing Pc resulted in an increase of the phase difference between HbO and Posc, Hbr remaining out of phase with Posc. These trends could be predicted with a lumped model of the forearm vasculature, suggesting that the technique could be used to asses mechanical characteristics of vascular beds.

We have compared in vivo spectra of the human forearm, brain, breast and blood-perfluorocarbon exchanged cat brain with reference spectra of deoxyhemoglobin, oxyhemoglobin, triglycerides, water and albumin to identify spectral features related to tissue biochemistry. From these investigations it is evident that bands due to fat, protein and water overlap the deoxyhemoglobin band at 760 nm. While these overlapping absorptions may be insignificant in whole blood, there is a potential for spectral interference in in vivo studies. Furthermore, unique features related the protein, lipids and water can be used to image breast tissue composition. The ability to resolve the overlapping absorption bands appears easily accomplished by using derivative mathematical approaches.

Near infrared spectroscopy was used to monitor changes in hemoglobin concentration in the kidney of 9 anesthetized rabbits during ischemic conditions. Laser diodes (775, 810, 862, and 904 nm) were placed on the surface of the kidney, the transmitted light was detected by a photomultiplier. Three stages of ischemia were induced in the kidney by occluding the abdominal aorta to approximately 1/3, 2/3 and total occlusion. Hemoglobin content was monitored during 10 minutes of occlusion followed by 10 minutes of recovery. The occlusions resulted in an average decrease in blood pressure of 36%, 53% and 100% in the abdominal aorta. The partial occlusions resulted in an increasing trend in total hemoglobin content during the 10 minutes occlusion followed by a return to baseline hemoglobin concentration. Within the first 30 seconds of the total occlusion maneuver, an immediate decrease in hemoglobin content was observed. During the occlusion the hemoglobin content increases slowly, corresponding to an increase in reduced hemoglobin content. Oxygenated hemoglobin content remained constant during the occlusion and increased immediately when the occlusion was released.

In this paper, we present a series of measurements made with a portable frequency-domain near-infrared tissue spectrometer (OMNIA). This is the first application of the OMNIA in a clinical setting. All of the measurements presented here were taken in vivo, most were on human subjects. We report the results of three experiments: (1) A simple ischemia/plethysmography experiment, which indicates ability of the instrument to noninvasively, continuously monitor the hemoglobin saturation of a limb. (2) A survey of hemoglobin saturation in patients with peripheral vascular disease. (3) An animal experiment to demonstrate the correlation of our instrument readings with results from established techniques for measuring hemoglobin saturation. We measured the absorption and reduced scattering coefficients of the tissue at two wavelengths (715 nm and 850 nm). From the absorption coefficients, we calculated the concentrations of oxygenated and deoxygenated hemoglobin ([HbO₂] and [Hb]), which immediately yield the hemoglobin saturation (Y) and the total blood volume (T) in the tissue. Our preliminary results indicate some of the potential of the instrument and the areas for future improvement of it.

Vastus lateralis muscle oxygenation was investigated on volunteers as well as muscular dystrophy patients during a walking test, and on volunteers during a free running by a continuous wave near infrared instrument. The data were analyzed using an oxygenation index independent on pathlength changes. Walking did not significantly affect the oxygenation of volunteers and patients. A relative deoxygenation was found only during free running indicating an unbalance between oxygen supply and tissue oxygen extraction. Preliminary measurements of exercising muscle oxygen saturation were performed by a 110 MHz frequency-domain, multisource instrument.

Rapid microlightguide spectrometers (EMPHO IIa/b) and a multiwire pO₂ electrode are applied for measurements of heterogeneous distribution of tissue oxygenation and redox state of respiratory enzymes in heart and rat liver. Optical and pO₂ measurements are noninvasively performed by use of sensors placed on the surface of tissue. Measurements in isolated perfused rat and in dog heart in situ were performed in order to investigate the relation between myocardial oxygenation and function. The tissue monitoring in liver was initiated by optical and polarographic monitoring in the hemoglobin free perfused organ. Subsequently, erythrocytes were added to the perfusate in several steps. The experiments reveal clear evidence that a protective system of tissue is activated when critical pO₂ values at the lethal corner of micro vessels fall off a critical threshold around 5 mmHg, thus causing a depletion of oxidative metabolism.

The ability to image and monitor the metabolic activity of keratocytes is important for the investigation of wound healing and repair mechanisms in the cornea. After laser refractive surgery there is activation of the stromal keratocytes in the human cornea. Two-photon excitation laser scanning microscopy was used to monitor the NAD(P)H levels in keratocytes in the cornea. The autofluorescence was confirmed to be mostly of NAD(P)H origin by treatment with cyanide which caused an increase in the fluorescence by a factor of two. We used a real-time scanning slit confocal microscope to image the distribution of keratocytes in the full thickness of the cornea. This microscope has the ability to image the cellular processes as well as the nuclei of the stromal keratocytes. Noninvasive optical imaging may provide a useful tool to investigate keratocyte activation after laser surgery or wound healing.

The processes which occur during interaction of radiation of erbium laser ((lambda) equals 2.9 micrometers ) and holmium one ((lambda) equals 2.1 micrometers ) with tissues have been studied recently considering medical applications. The phenomena of generation of intensive shock wave during propagation of Er laser radiation in water are well known. The relations of absorption coefficient (alpha) as a function of wavelength, presented in literature, show the essential differences of its values for these lasers; (alpha) equals 80 cm^-1 for Ho laser and (alpha) equals 13000 cm^-1 for Er laser. The calculations of volumetric density of radiation energy in a tissue have been made. The values of energy density obtained for free generation of both Ho and Er lasers exceed of several orders the energy radiation density of Nd:YAG laser with the narrowed pulse. In such a case the interactions of radiation of both lasers Er and Ho, respectively, will have significant nonlinear effects. The experiments verifying a value of radiation transmission of Er and Ho lasers, in water, have been carried out. A thickness of water layer is a variable parameter. For the same energy density of laser radiation the threshold value, in excess of which occurs the measurable value of energy at the output of experiment set-up, is significantly different in case of both lasers radiation. The value of threshold energy and transmission for energy density above the threshold have been measured. The results of experiment confirmed that Er laser radiation causes the stronger destruction in the deeper layers of the tissues. The experimental verification has been made at the anatomical specimens.