Hardware for obtaining spectral image cubes using filters or interferometers, though capable of revealing subtle aspects of composition, is typically expensive, bulky, slow, and often provides poor spatial resolution. In addition, a great deal of computer processing is needed to extract information from the raw data: interferometers must perform FFTs on megapixel data sets; all approaches involve calculations of spectral indices in order to classify or analyze the scene into its components. Consequently, spectral imaging techniques have been adopted only be a small, pioneering community. We report on a novel agile lamp for imaging which produces illumination having any desired spectral flux distribution ranging from pure spectral bands to precisely-tailored, complex polychromatic functions. This lamp, together with a CCD camera, is suitable for use in most spectral imaging applications, and by enabling one to directly image the scene in the spectral measure of interest, it eliminates the need for computer processing. Using matched filtering, one can obtain full information from all spectral bands in a handful of exposures with optimal signal-to-noise. The lamp is long-lived and spectrally stable. It will be affordable, compact, and rugged, and its spectral output can be adjusted within one millisecond. Finally, as there is no interferometer or other optics involved, imaging is 100% photon-efficient. Use of spectrally agile lamps in various multispectral applications is expected.

A high-performance hyperspectral imaging module with high light throughput suitable for microscopy and analytical imaging was built and tested. The imager utilizes the phenomenon of optical activity. The new technique provides information from a continuous spectral range of 250 - 1000nm. Similar spectral range extended to the near IR is also achievable. The imager has the form of a small module which can be inserted between a microscope or other imaging system and a camera. We have tested an 8-bit CCD video-rate camera with satisfactory results. The resulting instrument is simple, robust, and highly compact. The imager module is placed in-line to the microscope imaging system and does not introduce observable image aberrations. The imager is transparent to conventional imaging operations, thus with the imager in-place there is no need for reconfiguration of the microscope or switching between conventional and hyperspectral video/digital imaging modes. The presented spectral imager answers the need for a sensitive, compact, and affordable imaging spectrometer. The instrument is suited for applications requiring parallel acquisition of highly resolved concurrent spatial and spectral information such as high throughput screening, biochip analysis, remote sensing, semiconductor testing, etc. Images, spectral maps, and spectra of various fluorescent objects are presented.

Recent papers comparing the relative signal to noise performance of filters and interferometers for visible and near-IR spectral imaging have reached divergent conclusions. Others have presented the proposition that interferometer systems, by capturing light from multiple spectral channels bands at once, inherently outperform filter-type systems, which capture a single channel at a time. In this paper, a general analysis is provided that establishes a basis for comparison between band-sequential spectrometers and imaging interferometers in the shot-noise limit (set by Poisson statistics of the number of electrons produced at the detector), and in practice. It is shown that factors such as lamp flicker and sample stability can introduce much more error than shot noise does. This, one must be aware that the use of shot-noise limited detectors does not ensure shot- noise limited images.

The human eyes are not made to detect disease, however visual perception is the most common screening method for early cancer detection. With optimal illumination and observation configuration there is significant improvement of optical contrast between normal and pre-cancerous tissue in the oral cavity, both for reflected and fluorescent light.

The Optical Sciences & Engineering Research Center (OSER) at Virginia Polytechnic and State University investigates advanced laser surgery optics, biocompatible material for implants, and diagnostic patches and other diagnostic and drug delivery tools. The Center employs optics to provide new biological research tools for visualization, measurement, analysis and manipulation. The Center's Research into Multispectral Medical Analysis and Visualization techniques will allow human and veterinary medical professionals to diagnose various conditions of the body in much the same way that satellite information is used to study earth resources. Each pixel in the image has an associated spectra. Advanced image analysis techniques are combined with cross-correlation of the spectra with signatures of known conditions, allowing automated diagnostic assistance to physicians. The analysis and visualization system consists of five components: data acquisition, data storage, data standardization, data analysis, and data visualization. OSER research efforts will be directed toward investigations of these system components as an integrated tool for next generation medical diagnostics. OSER will research critical data quality and data storage issues, mult-spectral sensor technologies, data analysis techniques, and diagnostic visualization systems including the VT-CAVE, (www.cave.vt.edu). The VT-CAVE is Virginia Tech's configuration of Fakespace Systems, Inc Virtual Reality system.

In this paper, we describe a technique whereby cellular automata are used to rapidly scan hyperspectral medical images and quantify the extent of conditions of medical interest. The cellular automata population uses the condition of interest as food and only grows in those areas of the image where the food is present. The size of the cellular automata population can be correlated with the fractional area of the image containing the condition of interest. The technique has the potential to significantly reduce the computational overhead required to analyze a hyperspectral image. A simple model of the technique will be described and the results of its operation on a specific hyperspectral image is presented.

A recurrent problem in the assessment of thermal injuries is the ability to accurately identify the depth and extent of injury. Generally, the depth of a burn injury determines and is inversely related to the ability of the skin to restore and regenerate itself. Burns involve damage to the dermis in varying amounts, reducing the dermal blood supply and altering the skin hemodynamics. Near infrared spectroscopic imaging was used to non-invasively assess the changes that occur in the early (1-3 h) post-burn period. The study used an accurate porcine model to investigate the potential of near infrared spectroscopic imaging to accurately distinguish between burns of varying severity. Data analysis was carried out using a two-way and three-way data decompositions techniques to investigate the spectral changes related to burns. Burn injuries drastically alter the physical and optical properties of the tissue. Thermal destruction of cutaneous vasculature disrupts perfusion and oxygen delivery to the affected tissue. The results demonstrated that near infrared spectroscopic imaging might provide a new tool for an objective clinical assessment of burn injuries.

We have applied Fourier transform infrared (FTIR) spectroscopic imaging, coupling a mercury cadmium telluride (MCT) focal plane array detector (FPA) and a Michelson step scan interferometer, to the investigation of various states of malignant human prostate tissue. The MCT FPA used consists of 64x64 pixels, each 61 micrometers ², and has a spectral range of 2-10.5 microns. Each imaging data set was collected at 16^-1 resolution, resulting in 512 image planes and a total of 4096 interferograms. In this article we describe a method for separating different tissue types contained within FTIR spectroscopic imaging data sets of human prostate tissue biopsies. We present images, generated by the Fuzzy C-Means clustering algorithm, which demonstrate the successful partitioning of distinct tissue type domains. Additionally, analysis of differences in the centroid spectra corresponding to different tissue types provides an insight into their biochemical composition. Lastly, we demonstrate the ability to partition tissue type regions in a different data set using centroid spectra calculated from the original data set. This has implications for the use of the Fuzzy C-Means algorithm as an automated technique for the separation and examination of tissue domains in biopsy samples.

From November 1, 1997 till November 1, 2000 we have investigated 204 cases of acute myeloid leukemia (AML) (nequals95), acute lymphatic leukemia (ALL) (nequals40), myelodysplastic syndrome (MDS) (nequals11), chronic myeloid leukemia (CML) (nequals9), chronic lymphatic leukemia (CLL) (nequals4) and non-Hodgkin lymphoma (NHL) (nequals45) cytogenetically, using G-band analysis and spectral karyotyping (SKY). By SKY we were able to detect the abnormal clones in all cases but 9. In the G-band preparations these cases showed very few abnormal mitoses. The SKY either extended or confirmed the G-band findings in 94% of those with an abnormal karyotype. Cryptic translocations (translocations not suspected from the G-band karyotype) were found in 71 cases (26 AML, 9 ALL, 5 MDS, 2 CLL and 29 NHL). We find SKY a powerful adjuvant diagnostic tool that does not compromise one of the advantages of karyotyping techniques, the analysis of the entire genome which, in contrast to molecular biological techniques, still leave the possibility to get mroe answers than questions posed.

Multi-color fluorescence imaging of tissue samples has been an urgent requirement in current biology. As far as fluorescence signals should be isolated by optical bandpass filter-sets, rareness of the combination of chromophores with little spectral overlap has hampered to satisfy this demand. Additivity of signals in a fluorescence image accepts applying linear unmixing of superposed spectra based on singular value decomposition, hence complete separation of the fluorescence signals fairly overlapping each other. We have developed 7-color fluorescence imaging based on this principle and applied the method to the investigation of mouse spleen. Not only rough structural features in a spleen such as red pulp, marginal zone, and white pulp, but also fine structures of them, periarteriolar lymphocyte sheath (PALS), follicle, and germinal center were clearly pictured simultaneously. The distributions of subsets of dendritic cells (DC) and macrophages (M(phi) ) markers such as BM8, F4/80, MOMA2 and Mac3 around the marginal zone were imagined simultaneously. Their inhomogeneous expressions were clearly demonstrated. These results show the usefulness of the method in the study of the structure that consists of many kinds of cells and in the identification of cells characterized by multiple markers.

Optical spectroscopy has been shown to be an effective method for detecting neoplasia of epithelial tissues. Most studies to date in this realm have applied fluorescence or reflectance spectroscopy alone as a preferred method of disease detection. We have been developing instrumentation which can acquire both reflectance and fluorescence images of the human cervix in vivo, with the goal of combining multispectral information from the two spectroscopic modalities. This instrumentation has been tested on a group of patients in a clinical setting. We have applied spectral and spatial analysis techniques to the acquired images to assess the capabilities of this technology to discriminate neoplastic from normal cervical tissue.

Skin hydration is a key factor in skin health. Hydration measurements can provide diagnostic information on the condition of skin and can indicate the integrity of the skin barrier function. Near-infrared spectroscopy measures the water content of living tissue by its effect on tissue reflectance at a particular wavelength. Imaging has the important advantage of showing the degree of hydration as a function of location. Short-wavelength (650-1050 nm) near infrared spectroscopic reflectance imaging has previously been used in-vivo to determine the relative water content of skin under carefully controlled laboratory conditions. We have recently developed a novel spectroscopic imaging system to acquire image sets in the long-wavelength region of the near infrared (960 to 1700 nm), where the water absorption bands are more intense. The LW-NIR systems uses a liquid- crystal tunable filter in front of the objective lens and incorporates a 12-bit digital camera with a 320-by-240-pixel indium-gallium arsenide array sensor. Custom software controls the camera and tunable filter, allowing image sets to be acquired and displayed in near-real time. Forearm skin hydration was measured in a clinical context using the long- wavelength imaging system, a short-wavelength imaging system, and non-imaging instrumentation. Among these, the LW-NIR system appears to be the most sensitive at measuring dehydration of skin.

Spectral imaging as a modality combines two powerful techniques of imaging and fluorescence spectroscopy. By generating a 2-D image of an object with fluorescence information at every pixel, spectral imaging has the potential to provide clinicians with a valuable tool which can not only diagnosis the tissue but also provide an image of the boundary of where the normal and cancerous tissue intersect. The system used in these experiments was a modified spectral imaging system from Applied Spectral Imaging (SD-200, Carlsbad, CA). The system was mounted in an off-microscope configuration so that instead of performing microscopic measurements of tissue, macroscopic measurements on the order of several millimeters in size were collected. Preliminary results indicate that the spectra acquired from human brain tissues in vitro at individual pixels of the spectral image cube appear similar to that acquired using the single pixel system. Based on the findings of this study, spectral imaging has the potential to be a useful tool for tissue diagnostics and is currently limited by the speed of data acquisition and size of the data.

As a precursor to applying fluorescence lifetime imaging (FLIM) to studies of intercellular communication in molecular immunology, we have investigated the fluorescence lifetime of enhanced green fluorescent protein (EGFP) in mixtures of water and glycerol using time-correlated single photon counting (TCSPC). We find that the EGFP lifetime decreases with increasing glycerol content. This is accounted for quantitatively by the refractive index dependence of the fluorescence lifetime as predicted by the Strickler Berg formula which relates the fluorescence lifetime to the absorption spectrum. The solvent viscosity has no influence on the fluorescence lifetime. We also discuss the refractive index dependence of the GFP fluorescence lifetime in more complex systems. The findings are particularly relevant for the interpretation of FLIM of GFP expressed in environments such as bacteria and cells.

Advances in the design and miniaturization of the lasers and electronics required for Time Correlated Single Photon Counting (TCSPC) measurement of fluorescence lifetime have simplified the use of the time domain method. We have assembled a compact portable system that is capable of measuring lifetimes down to approximately 200 ps (with deconvolution) and that can operate at a range of excitation and emission wavelengths. The excitation sources are pulsed LEDs and laser diodes with a maximum pulse rate of 40 MHz and are easily interchanged. Furthermore, the development of violet and blue GaN LEDs and laser diodes is expanding the range of fluorophores available for fluorescence lifetime measurement of ion concentrations. pH sensitive fluorophores have a wide range of biological and clinical applications. The use of fluorescence lifetime rather than intensity to measure pH has a number of advantages including the reduction of effects due to the photobleaching, scattering, and intensity variations in the excitation source. Using our compact TCSPC instrumentation we have measured the dependence of fluorescence lifetimes on pH for a range of dyes in phosphate buffer over the physiologically important range of 6.0 to 8.0. Most dyes exhibit only a small variation in lifetime (<1.0ns) over the 6.0 to 8.0 pH range; however, acridine exhibits a large variation in lifetime and hence shows promise as a pH indicator.

The spatiotemporal properties of the Ca²⁺ release process in skeletal muscle fibers were determined using an improved confocal spot detection system. Muscle fibers were loaded with the low affinity fluorescent Ca²⁺ indicator OGB-5N and localized action potential-induced fluorescence signals were recorded from consecutive locations separated by 200 nm within a single sarcomere. Three-dimensional reconstructions on the Ca²⁺ transients illustrate the existence of domains of increased fluorescence around Ca²⁺ release sites in the neighborhood of the T-tubules. We estimated the dimensions of these domains by drawing isochronal curves ((delta) F/F vs. spot position) and fitting Gaussian profiles to them. It was found that the earliest detectable full-width-at-half- maximum of these profiles was 0.77 +/- 0.25 micrometers and increased rapidly with time to 1.4 +/- 0.2 micrometers at their peak (18 degree(s)C). A brief, but statistically significant delay of 0.8 +/- 0.42ms was observed between the onset of the fluorescent transients at the Z- and M-lines. Our results are compatible with the possibility that, in response to AP stimulation, Ca²⁺ is not released exclusively from the junctional region of the sarcoplasmic reticulum, but from a broader expanse of the triadic region.

Site specific delivery of drugs and contrast agents to tumors protects normal tissues from the cytotoxic effect of drugs, and enhances the contrast between normal and diseased tissues. In optical medicine, biocompatible dyes can be used as phototherapeutics or as contrast agents. Previous studies have shown that the use of covalent or non-covalent dye conjugates of carriers such as antibiodies, liposomes, and polysaccharides improves the delivery of such molecules to tumors. However, large biomolecules can elicit adverse immunogenic reactions and also result in long blood clearance times, delaying visualization of target tissues. A viable alternative to this strategy is to use small bioactive molecule-dye conjugates. These molecules have several advantages over large biomolecules, including ease of synthesis of a variety of high purity compounds for combinatorial screening of new targets, enhanced diffusivity to solid tumors, and the ability to affect the pharmacokinetics of the conjugates by minor structural changes. Thus, we conjugated a near infrared absorbing dye to several bioactive peptides that specifically target overexpressed tumor receptors in established rat tumor lines. High tumor uptake of the conjugates was obtained without loss of either the peptide receptor affinity or the dye fluorescence. These findings demonstrate the efficacy of a small peptide-dye conjugate strategy for in vivo tumor imaging. Site-specific delivery of photodynamic therapy agents may also benefit from this approach.

Oral cancer and precancer overexpress the epidermal growth factor receptor (EGFR) and monoclonal antibodies against EGFR coupled to photoactive dyes may have a potential both as a diagnostic and treatment modalities for oral premalignancy. We asked whether an anti-EGFR mab (C225) conjugated with the fluorescence dye indocyanine Cy5.5 could detect dysplastic changes in the hamster cheek pouch carcinogenesis model. Secondly, we tested whether the same antibody conjugated with the photosensitizer chlorin (e6) could be used together with illumination to reduce levels of expression of EGFR as evaluated by the immunophotodetection procedure. Increased fluorescence appeared to correlate with development of premalignancy when the C225-Cy5.5 conjugate was used. Areas with increased fluorescence signal were found in carcinogen-treated but clinically normal cheek pouches, that revealed dysplastsic changes by histology. The immunophotodetection procedure was carried out after photoummunotherapy with the C225-ce6 conjugate, and showed a significant reduction in fluorescence in the illuminated compared to the non-illuminated areas in the carcinogen- treated but not the normal cheek pouch. The results demonstrate that the use of anti-EGFR Mab targeted photoactive dyes may serve as a feedback controlled optical diagnosis and therapy procedure for oral premalignant lesions.

Our understanding of intracellular pH homeostatis in eukaryotic systems has been enhanced since the introduction of carboxyfluorescein diacetate as a useful pH probe more than 20 years ago. BCECF, a derivative of this earlier fluoroprobe has dominated the field. In the past 10 years, SNARF-1 has emerged as an alternative pH probe. Recently, a novel derivative of BCECF, BCPCF has been developed. Green Fluorescent Proteins (GFPs) have also been used recently to monitor pH in a non invasive manner in several cell types. Here, we report that human mammary epithelial cells can be transfected with the gene encoding for cyan (CFP), green (GFP), and yellow (YFP), to study cytosolic pH. The novel red fluorescent protein (DsRed) is not sensitive to pH. Multidrug resistance (MDR) has been associated with altered cytosolic pH homeostasis. We show that experimental maneuvers that decrease pHⁱⁿ enhance the efficacy of chemotherapeutic drugs. We also show that short pulses of UV-B light elicited acidosis in cells, as evaluated by ratio ion cell imaging, and confocal/spectral imaging microscopy. During the course of these experiments we noticed that cells exhibit intrinsic fluorochromes that can be used to monitor pH in living cells.

Indocyanide green (ICG) is widely used as a tracer for the non-invasive estimation of liver function. ICG has properties of binding with plasma protein, and has a large absorption peak at 805 nm. There were no reports, however, about the IR absorption peak of ICG at 7.1 micrometers , which absorption coefficient amounts to approximately 13000cm^-1. In this study, ICG was exposed to free electron lasers (FELs) with wavelength of 7.1 micrometers and usefulness of ICG as an IR-marker was discussed. ICG film sample was formed on IR-transparent BaF₂ crystal substrate and exposed to FELs with the wavelength of 7.1micrometers . After exposure the sample was analyzed by FT-IR and film thickness measurements. As results, ICG ablated with the FEL of the power density of more than 5 W/cm²(equalsP_dth), and that the molecular structure of ICG was still stable for the power density of less than P_dth, 3 W/cm². Therefore, ICG can be considered as a novel infrared marker (IR marker) to the living tissue which absorbs FEL photon energy without changing the IR absorption peak.

Five peptides binding to somatostatin and bombesin receptors were conjugated to 4-azido-2,3,4,6-tetrafluorophenylbenzoic acid, a Type 1 photosensitizer, at the N-terminal position. The receptor affinities were determined by competition binding assay using two different pancreatic tumor cell lines, CA20948 and AR42-J, that expresses somatostatin-2 (SST-2) and bombesin receptors receptively. All compounds exhibited high receptor specificity, i.e., the IC₅₀ values ranged between 1.0 to 64.0 nM. These conjugates may be useful for targeted Type 1 phototherapy via the generation of nitrenes at the cell surfaces expressing these receptors.