We describe a novel three-beam confocal laser Doppler microscope (CLDM) which allows 2D microscopic velocity flow measurements and utilizes the benefits of both the Doppler mode of contrast formation and the high lateral and axial resolution imaging capability of confocal microscopy. This is accomplished by a microscope objective focusing three laser beams to produce a 'probe volume' within a specimen, rather than the traditional two input beams. The specimen is mounted on a scanning stage, which enables the final image to be constructed point by point. The two dimensional velocity information of the particles within this probe volume are 'encoded' into the beat frequencies of the back scattered light. By using digital filtering to deinterlace the individual signals and computer reconstruction techniques, 'velocity images' will be formed. These velocity images, produced solely by Doppler scattering of light from moving particles, displays the flow distribution within some defined area. Our goal is to eventually produce a full three dimensional velocity volume of a biological cell, which would truly be a significant innovation in microscopy. We demonstrate an experimental optical device capable of forming the required three independent fringe fields inside the probe volume, examine this probe volume from a theoretical basis and perform an analysis of the scattered light frequencies. Preliminary experimental results are presented.

We have built an interference microscope that produces in real-time images of cross-section slices located at adjustable depths inside 3-D objects. The microscope is based on a Michelson-type polarization interferometer. A light emitting diode (LED), used as an optical source at (lambda) equals 840 nm with short coherence length, provides optical sectioning ability with better than 10 micrometer resolution in the depth dimension. By using high numerical aperture objective lenses (NA equals 0.95), the depth resolution can be improved to better than 1 micrometer, in good agreement with theory. Images can be produced at the rate of 50 per second using a multiplexed lock-in detection and MMX assembler-optimized calculation routines. Cross-section images inside an onion and at different depths in a multilayer silicon integrated circuit are presented.

An important approach dealing with sub lambda super resolution effect is based on a 3-D scanning of the examined sample using a pipette with small diameter. The scanning in the z direction is required for the conservation of the distance between the sample and the pipette and it is time consuming. In this paper we suggest a modification of that approach based on 2-D scanning. The scanning in the z direction is replaced by the usage of white light illumination. Based on the spectrum of the scanned information, one can compute the height of the surface. This consumes much less time than the conventional pipette scanner. This approach appears to be reasonable if the profile is varying not more than a portion of a wavelength. If, however, the object is not flat but slightly curved on a large scale we propose to employ a servo for the height. This servo can be much slower than the signal servo in the traditional approach.

Microscopy based on the mid-infrared part of the spectrum provides an approach to imaging with a chemically selective contrast mechanism. However, the long wavelength of the mid-IR radiation diffraction limits the spatial resolution to no better than a few microns in conventional IR microscopy. (In practice, commercial IR microscopes rarely do as well as 10 microns). This resolution prevents IR spectroscopy of single sub-cellular or even cellular features routinely observable with conventional visible light microscopes. In this paper we present a technique using conventional IR optics and no near- field tips which enables IR imaging with the resolution of a visible microscope. Photo-induced reflectivity generated by ps pulses of visible light incident on the surface of semiconductor is used to create a transient mirror with dimensions determined by the spot size of the visible light. The IR light scattered by such subwavelength-size mirror is collected after propagating through the sample. As the sample is located on the semiconductor substrate, no near-field distance control is required, and the image can be taken at the speed of a typical laser scanning microscope. And since the near-field probe is generated remotely -- using light -- the sample to be imaged can be covered by, or encased in, a transparent liquid or solid. The resolution of such an IR microscope is determined by the dimensions of the transient mirror, i.e. by the spot size of the visible light and its penetration depth into the substrate. To prevent resolution degradation due to diffusion of the photo-excited carriers in the substrate, the probe (IR) pulse duration should not exceed a few tens of picoseconds. Preliminary results, prospects and limitations are discussed.

High intensity chirped pulses can be used for probing microscopic chemical environments through the use of a particular choice of dye, for instance SNAFL2. The basis for this technique is that the excited state populations can be manipulated through control over the temporal order of the excitation frequencies in the excitation pulse -- i.e. chirp - - with the outcoming fluorescence as the reporting parameter. A chirp dependent fluorescence response can also be observed in larger molecular systems with more degrees of freedom like for instance green fluorescent proteins. In preparation for application of the technique to microscopy we use a facility permitting observation of this phenomenon in various dyes with high sensitivity. High power, 30 fs pulses from an OPA, tunable from 400 nm to 1.5 micron are used. These pulses with a repetition rate of 1 kHz are sufficiently intense that a relatively large sample region can be excited to saturation from which then a sub-region with uniform excitation conditions can be selected for signal collection.

Confocal theta microscopy improves the resolution of confocal laser scanning microscopes by solving the problem of the inferior axial resolution instrumentally. The specimens are observed at an angle theta relative to the illumination axis (ideally 90 degrees). The resulting observation volume, defined by the product of illumination and detection point spread functions (PSFs), is reduced by 2.5 and has an isotropic shape. Single-Lens Theta Microscopy (SLTM) is a technical variation of confocal theta microscopy. It is designed to be easily adapted to any common confocal laser scanning microscope. It is based on the use of a mirror unit between the microscope objective lens and its focal plane. This mirror unit deflects the incoming and outcoming light in such a way, that the detection axis is perpendicular to the illumination axis. With SLTM different kinds of other microscopical techniques (such as 4Pi microscopy) are possible. The quantitative evaluation of physical test systems underline the feasibility of SLTM and prove the excellent resolution. The extensions of the experimentally determined point spread functions fit well with the predicted theoretical values. The technique was applied to the investigation of GFP labelled organelles in HeLa cells as well as for the analysis of embryos.

One of the promising recent developments in fluorescence microscopy is fluorescence lifetime imaging microscopy (FLIM). In this technique the fluorescence lifetime (ns range) of molecules is expressed in the image rather than the intensity of the light emitted by these molecules. This physical property is of interest as it gives information about the local environment of the molecule, such as molecular concentration of O₂, Ca²⁺, pH, and conjugation. We develop an affordable, robust and easy-to-use FLIM workstation which is completely automated and does not need any difficult calibration procedure. The system consists of a standard fluorescence microscope, a modulated excitation light source, a camera, a modulation signal generator and acquisition/processing software. The camera contains an Intensified CCD of which the image intensifier gain is modulated. Depending on the application different light sources can be selected. The current light source contains a 12 mW modulated laser-diode emitting at 635 nm. A homodyne detection scheme with modulation frequencies of 1 to 100 MHz is applied, aiming at a resolution of 0.1 ns or better. High level image acquisition strategies are implemented in software, along with the low level image processing routines for lifetime estimation, calibration and correction. An evaluation of the system and its critical components will be presented in this paper.

Confocal photoluminescence imaging is an important tool in the investigation of recombination in semiconductors and in the characterization of material growth. This characterization is particularly important for II-VI wide band-gap semiconductors where the potential for blue-green lasers is being explored currently. To achieve room-temperature cw operation of these lasers over the multi-thousand hours necessary for commercialization, extremely low defect densities are required. The confocal microscope is used in this work to image photoluminescence from II-VI materials to characterize the defect formation and propagation within the quantum well region of the material. This imaging approach permits the degradation to be monitored in real time and over a large area in samples with low defect densities. The additional advantages of this set-up over a conventional microscope are, of course, the higher lateral resolution and narrow depth of field associated with a confocal microscope. While considerable effort has been focused on the degradation in these II-VI semiconductors, we have recently observed that annealing can occur simultaneously in the same sample when the material is exposed to intense optical excitation. Images of annealing and degradation of a range of II-VI samples will be presented to highlight these observations.

A new generation of confocal laser microscope, designed to image the human skin in vivo, improves the resolution, contrast and spectroscopic facilities as compared to the previous Tandem Scanning Microscope (TSM) prototype. The new device has been built with an Oz module (Noran) equipped with the skin contact device, assuming a perfect stability of skin images in the horizontal plane. The Z displacement of the objective lens, mounted directly on the Oz module, is assumed by a piezo motor with a course of 350 micrometer. Moreover, the Oz module has been suspended on articulated arms to reach any part of the human body. The power of the Argon/Krypton laser source has been limited to 2.mW to secure safety and provides three visible wavelength: 488, 568, and 647 nm. The facility of instantly checking wavelength during in depth exploration of the skin optimizes the resolution and contrast of images as compared with the white light used in the TSM. Consequently, better image quality of the epidermis is obtained in the blue region with unexpected details of corneocytes and keratinocytes. The papillary dermis comprising the vascular network is advantageously observed with the red light. The fluorescence channel detector gives additional information on the penetration of fluorescent probes through the skin barrier. Optical sections are digitized (512 X 480 X 8 bit) at video rate, providing easy and fast measurements of the thickness of epidermal layers. The Silicon Graphics workstation generates a transparent volume of living human skin in less than 5 minutes. This powerful and convivial new design for imaging the in vivo human skin opens up new promises in skin research and in vivo skin pharmacology.

Recently a novel imaging technique based on third-harmonic generation (THG) was introduced. This technique relies on a third-order non-linear interaction to generate a coherent signal response on the third-harmonic frequency with respect to the fundamental input radiation. Here we report on the input NA dependence of the THG signal and examine the resulting imaging characteristics of this novel technique in terms of resolution and contrast generation. We'll demonstrate the potential of the technique through a number of imaging examples, with special emphasis on in vivo applications. The latter illustrates the non-invasive character of the technique.

The optical quadrature imaging technique, as derived and extended from microwave and laser radar quadrature detection techniques, provides an efficient method for obtaining phase information from a sample that has little or no amplitude contrast. We are able to resolve internal structures of a sample that are defined by relatively small refractive index differences without the use of dyes or stains, while using much lower light levels than conventional techniques. We have constructed a prototype system for imaging microscopic phase- only objects. In this paper, we present its capabilities, as well as the imaging and reconstruction methods used to obtain quantitative information about a sample.

We present a novel light efficient technique for obtaining optically sectioned images in wide-field microscopy. The technique is a further development of correlation microscopy and is based on using complementary structured light patterns to illuminate the specimen together with uncomplicated post- processing of the captured images.

The spatial point spread function is studied for two-photon fluorescence imaging in turbid media. Using a Monte Carlo based model the effects of tissue optical properties on the point spread function, fluorescence generation, and fluorescence collection are studied. The simulated results are compared to measured point spread functions at different depths within a scattering sample. Results indicate that the limiting factor in two-photon imaging is a loss of signal caused by scattering rather than a loss of resolution.

Calibration of the axial response of interference microscopes has received considerable attention in the past two decades. In addition to systematic errors which could be caused by components in the microscope or measurement technique, a numerical correction factor associated with imaging at high apertures must be determined. Unfortunately, the cost of reference height standards increases sharply with their spatial homogeneity and calibration accuracy and these standards may be easily contaminated and therefore require sophisticated cleaning and re-calibration. To address these problems, we have investigated the interferometric measurement of the equilibrium shape of static fluid drops on coated substrates. For drops with small Bond number (the Bond number is a ratio of gravitational to capillary forces), the surface of the drop forms a spherical cap. It appears that nature forms a highly smooth, curved surface. By varying the surface energy, it is possible to obtain a wide range of static contact angles. For example, silicone oil [polydimethylsiloxane (PDMS)] on glass forms a contact angle of about 5 degrees, while it forms an angle of 38 degrees on Teflon and 68 degrees on a fluorinated silicon surface. We have measured contact angles as large as 68 degrees for PDMS on a single crystal silicon wafer with a 50 X/0.8 NA objective using a custom-made phase-shifted, laser feedback microscope. The method for preparing these static drops is simple and we envision that microscopists will be able to prepare easily disposable calibration standards in their laboratories.

We have studied three-dimensional reconstruction methods to estimate the cell volume of astroglial cells in primary culture. The studies are based on fluorescence imaging and optical sectioning. An automated image-acquisition system was developed to collect two-dimensional microscopic images. Images were reconstructed by the Linear Maximum a Posteriori method and the non-linear Maximum Likelihood Expectation Maximization (ML-EM) method. In addition, because of the high computational demand of the ML-EM algorithm, we have developed a fast variant of this method. (1) Advanced image analysis techniques were applied for accurate and automated cell volume determination. (2) The sensitivity and accuracy of the reconstruction methods were evaluated by using fluorescent micro-beads with known diameter. The algorithms were applied to fura-2-labeled astroglial cells in primary culture exposed to hypo- or hyper-osmotic stress. The results showed that the ML-EM reconstructed images are adequate for the determination of volume changes in cells or parts thereof.

Image compression in microscopy is a valuable technique, in particular if applied to multidimensional data. Information- preserving and competitive information-losing compression is applied to microscopical data and the resulting image quality is evaluated on a quantitative basis both in the spatial and frequency domain. Included are image data featuring different signal-to-noise ratios, but also voxel data for volume representations as well as graphical data for surface representations. The effect of compression on 3-D visualization and image management with data bases is included.

Cell volume changes are often associated with important physiological and pathological processes in the cell. These changes may be the means by which the cell interacts with its surrounding. Astroglial cells change their volume and shape under several circumstances that affect the central nervous system. Following an incidence of brain damage, such as a stroke or a traumatic brain injury, one of the first events seen is swelling of the astroglial cells. In order to study this and other similar phenomena, it is desirable to develop technical instrumentation and analysis methods capable of detecting and characterizing dynamic cell shape changes in a quantitative and robust way. We have developed a technique to monitor and to quantify the spatial and temporal volume changes in a single cell in primary culture. The technique is based on two- and three-dimensional fluorescence imaging. The temporal information is obtained from a sequence of microscope images, which are analyzed in real time. The spatial data is collected in a sequence of images from the microscope, which is automatically focused up and down through the specimen. The analysis of spatial data is performed off-line and consists of photobleaching compensation, focus restoration, filtering, segmentation and spatial volume estimation.

The essential difference of non-linear image restoration algorithms with linear image restoration filters is their capability to restrict the restoration result to non-negative intensities. The iterative constrained Tikhonov-Miller algorithm (ICTM) algorithm, for example, incorporates the non- negativity constraint by clipping each iteration of its conjugate gradient descent algorithm. This constraint will only be effective when the restored intensities have near zero values. Therefore the background estimation will have an influence on the effectiveness of the non-negativity constraint of these non-linear restoration algorithms. We have investigated the effect of the background estimation on the performance of the ICTM, Carrington, and Richardson-Lucy algorithms and compared it to the performance of the linear Tikhonov-Miller restoration filter. We found that an underestimation of the background will make the non-negativity constraint ineffective which results in a performance that does not differ much from the performance obtained by the linear restoration filter. An overestimation of the background however is even more dramatic since it results in a clipping of object intensities. We show that this will dramatically deteriorate the performance of the non-linear restoration algorithms. We propose a novel method to estimate the background based on the dependency of non-linear restoration algorithms on the background.

Objective: The application of an electronic slide and a software simulated virtual microscope can contribute to a more efficient, convenient histological analysis. These techniques would also support the automation of histological analysis and three dimensional reconstruction of histological objects. Study Design: A fully computer controlled microscope (Axioplan 2 MOT), video camera (Grundig FAC 830) and an Intel Pentium II based PC were used for the development of the electronic slide. The applied frame grabber had 640 X 560 resolution, 64 kb color depth. A program was developed, called Pyramid, for the scanning of an entire slide. Autofocusing, image reduction and frame joining algorithms were implemented in the virtual microscope application. Results: The autofocusing and digitization of one image segment (400X magnification, 0,0725 mm²) took 8 seconds. The harddisk volume of one frame is between 60 and 100 kilobytes (kb) after JPEG compression. The overall harddisk place for a gastric biopsy is around 130 - 150 megabytes. The Pyramid program contains routines for electronic evaluation of the slide. The major microscopic functions are implemented: moving in any free directions in discrete or continuous steps, magnifications on a discrete scale (400, 200, 100X), and in continuous scale. Up to 1o notes can be placed on any place of the slide and can be retrieved within a second. The program can be used in local area networks for slide evaluations. Conclusions: The scanning speed is now too low for routine application, however with further developments in data storage and imaging technology, the electronic slide and the virtual microscope can be alternative techniques in the computerization of the histology laboratory. After the scanning of consecutive sections and a mathematical matching procedure supracellular organizations from gastric biopsies were reconstructed giving new insight into tumor growth.

New method of a tomographic microscopy with spatially incoherent illumination is proposed. Earlier, in our works for reconstruction of a refractive index with spatial distribution of 3D phase samples, a method of tomographic microscopy with the coherent light was used. Main disadvantage of this scheme is use of a coherent light, which brings to appearance the phase noise in projection data. Spatially incoherent tomographic microscope is based on the Linnik phase-shifting interference microscope reflected type. Using interferometry with large source of the monochromatic light allows greatly improve a quality of phase projections by averaging of interference patterns. For tomographic mode of Linnik microscope working and obtaining the oblique illumination of sample method for displacing a light source is used. Observation angle range for 100 X oil-immersion objective, N.A. equals 1.25, is 90 degrees. This method allows realizing a two-dimensional scanning trajectory for sample observation. Particularities of tomography for the phase objects, placed near plane mirror, are considered. So, a viewing angle range can be reduced to 90 degree, and sample is complement by its mirror reflection. The iterative algorithms for limited-angle tomographic reconstruction were used. Experimental results on three-dimensional reconstruction of the single human blood cell (erythrocyte) are presented.

A major goal in neuroanatomy is to obtain precise information about the functional organization of neuronal assemblies and their interconnections. Therefore, the analysis of histological sections frequently requires high resolution images in combination with an overview about the structure. To overcome this conflict we have previously introduced a software for the automatic acquisition of multiple image stacks (3D-MISA) in confocal laser scanning microscopy. Here, we describe a Windows NT based software for fast and easy navigation through the multiple images stacks (MIS-browser), the visualization of individual channels and layers and the selection of user defined subregions. In addition, the MIS browser provides useful tools for the visualization and evaluation of the datavolume, as for instance brightness and contrast corrections of individual layers and channels. Moreover, it includes a maximum intensity projection, panning and zoom in/out functions within selected channels or focal planes (x/y) and tracking along the z-axis. The import module accepts any tiff-format and reconstructs the original image arrangement after the user has defined the sequence of images in x/y and z and the number of channels. The implemented export module allows storage of user defined subregions (new single image stacks) for further 3D-reconstruction and evaluation.

Confocal fluorescence imaging is widely used, particularly for biological applications, and also notably in direct-view microscopes. Recent work has compared the use of coherent and incoherent illumination sources on the optical sectioning characteristics of fluorescence direct-view microscopes. However this detailed comparison has been done in theory. This paper addresses the experimental aspects of using coherent light sources in fluorescence imaging using a range of finite- sized, multiple-aperture arrays. The experimental difficulties of choosing a suitable uniform, flat, fluorescent plane with a high quantum efficiency are considered. Axial response curves obtained with a fluorescent laser dye sample are presented.

Spectral imaging permits two-dimensional mapping of the reflectance properties of biological systems. However, imaging in turbid media involves pixel sizes that are comparable to or smaller than the mean photon path length. This implies that the spectrum measured at a given pixel in the image plane will be determined by manifold photon trajectories through an extended volume in the object, so there is not a uniquely defined path length. In addition, this implies nonlinear spectral mixing for systems with multiple layers and chromophores. Using Monte Carlo model, we have studied photon path distributions in the case of layered turbid systems and their effects on spectral imaging. In particular, we emphasize the effect of hemoglobin on imaging reflectance-mode hyperspectral data.

Hyperspectral imaging has traditionally required sophisticated software and powerful computers. In this paper, we describe an elegant PC-based system that maps multiple fluorescent signatures simultaneously. We present color image examples of the spectral topography of biological materials that exhibit as many as 40 distinct spectral fingerprints. Acquisitions and data processing occur close to real time. Unique coding simplifies memory management and data storage functions and provides accelerated image retrieval and interpretation.

The object of the experiment described in this paper was to demonstrate that cells stained with multiple fluorophores could be identified and quantified simultaneously. Hyperspectral imaging was used to classify spleen cells of a Balb/c mouse, with the anti-mouse CD4 antibody conjugated with Alexa 488, 532, 546 and 568. It was found that the system was able to identify the specific fluorophore present and map their location in the cells. The system also provided relative signal strength data. Spectral libraries were constructed with color-coded spectra that enabled automatic spectral identification in subsequent acquisitions.

Evaluation of cell morphology by bright field microscopy is the pillar of histopathological diagnosis. The need for quantitative and objective parameters for diagnosis has given rise to the development of morphometric methods. The development of spectral imaging for biological and medical applications introduced both fields to large amounts of information extracted from a single image. Spectroscopic analysis is based on the ability of a stained histological specimen to absorb, reflect, or emit photons in ways characteristic to its interactions with specific dyes. Spectral information obtained from a histological specimen is stored in a cube whose appellate signifies the two spatial dimensions of a flat sample (x and y) and the third dimension, the spectrum, representing the light intensity for every wavelength. The spectral information stored in the cube can be further processed by morphometric analysis and quantitative procedures. Such a procedure is spectral-similarity mapping (SSM), which enables the demarcation of areas occupied by the same type of material. SSM constructs new images of the specimen, revealing areas with similar stain-macromolecule characteristics and enhancing subcellular features. Spectral imaging combined with SSM reveals nuclear organization through the differentiation stages as well as in apoptotic and necrotic conditions and identifies specifically the nucleoli domains.

Transmittance spectra of optical micrograph of fungi were estimated from seven band images based on the Wiener estimation method. The optical microscope adjusted with monochrome digital camera (Kodak DCS420m, 1536 X 1024 pixels) with seven band filters was used for image acquisition. Transmittance spectra of 16 color transparencies were measured by spectral photometer in advance, and the seven band images of color transparency were taken by microscopic imaging system. A matrix for Wiener estimation was calculated using these measured transmittance spectra and camera responses. As the result of Wiener estimation, normalized root mean square error (NRMSE) between the original and estimated transmittance spectra of 16 color transparencies was 0.035. Wiener estimation method was applied to estimate the spectral transmittance of five species which belong to one genus of fungi. Transmittance spectra of fungi were calculated from camera responses in each pixel and the above estimation matrix. The estimated transmittance spectra were used for segmentation of conidia and hyphae in fungal microscopic image to identify them. Competitive learning in neural network was applied to the segmentation from spectral microscopic image. It was found that segmentation based on spectral transmittance was more accurate than the segmentation based on RGB values.

Confined arrays of biochemical probes deposited on a solid support surface (analytical microarray or 'chip') provide an opportunity to analysis multiple reactions simultaneously. Microarrays are increasingly used in genetics, medicine and environment scanning as research and analytical instruments. A power of microarray technology comes from its parallelism which grows with array miniaturization, minimization of reagent volume per reaction site and reaction multiplexing. An optical detector of microarray signals should combine high sensitivity, spatial and spectral resolution. Additionally, low-cost and a high processing rate are needed to transfer microarray technology into biomedical practice. We designed an imager that provides confocal and complete spectrum detection of entire fluorescently-labeled microarray in parallel. Imager uses microlens array, non-slit spectral decomposer, and high- sensitive detector (cooled CCD). Two imaging channels provide a simultaneous detection of localization, integrated and spectral intensities for each reaction site in microarray. A dimensional matching between microarray and imager's optics eliminates all in moving parts in instrumentation, enabling highly informative, fast and low-cost microarray detection. We report theory of confocal hyperspectral imaging with microlenses array and experimental data for implementation of developed imager to detect fluorescently labeled microarray with a density approximately 10³ sites per cm².

The structural elucidation of complex systems may be simplified with multi-dimensional spectroscopic techniques with some combination of spatial and spectral resolution. Raman spectroscopy permits the addition of another variable to this scenario -- excitation wavelength. Data obtained using excitation wavelengths from the UV (244 nm) to near-IR (785 nm) regions will be presented showing the qualitative and quantitative study of diamond-like carbon (DLC), silicon, and other systems of an industrial or biomedical nature. The choice of appropriate wavelength provides an additional advantage over other spectroscopic techniques for elucidating specific structural information from these systems. The advantages of UV-Raman for materials science and thin film studies will be considered. The design of instruments and probes for the application of Raman spectroscopy to industrial process control and the development of Raman spectroscopic libraries for contaminant analysis will be discussed.

The basic procedure of spectral image analysis is mapping, i.e., generation of conventional images using specific spectral features from multidimensional data set. Commonly, it is based on decomposition of the original spectra into a sum of model spectra which are usually attributed to chemically/physically different components. However, the problem is complicated when the model spectra are not clearly definable or difficult and even impossible to obtain experimentally. Here we describe a method for estimating spectral models based on the elimination of physical meaningless solutions obtained by PCA (Principal Component Analysis). The method includes the following steps: segmentation of the original spectral image using a white- light image; generation of primary factors for each segmented region by MDF (maximum distance factors) algorithm; extending the factor space to satisfy non-negative score requirements; generation of model limits by rotation of factor matrix in conditions of non-negative intensity values; calculation of the common model as intersection regions in the score space. The precision of the method depends on the spectral variations in the original data. When these variations are significant the models could be obtained precisely and then attributed to possible sample components. This method was validated in different research fields: to study the distribution of anticancer drugs in single living cells, to characterize phenolic species in wheat walls.

Radiocarcinogenesis is widely recognized as occupational, environmental and therapeutical hazard, but the underlying mechanisms and cellular targets have not yet been identified. We applied SKY to study chromosomal rearrangements leading to malignant transformation of irradiated thyroid epithelial cells. SKY is a recently developed technique to detect translocations involving non-homologous based on unique staining of all 24 human chromosomes by hybridization with a mixture of whole chromosome painting probes. A tuneable interferometer mounted on a fluorescence microscope in front of a CCD camera allows to record the 400 nm - 1000 nm fluorescence spectrum for each pixel in the image. After background correction, spectra recorded for each pixel are compared to reference spectra stored previously for each chromosome-specific probe. Thus, pixel spectra can be associated with specific chromosomes and displayed in 'classification' colors, which are defined so that even small translocations become readily discernible. SKY analysis was performed on several radiation-transformed cell lines. Line S48T was generated from a primary tumor of a child exposed to elevated levels of radiation following the Chernobyl nuclear accident. Subclones were generated from the human thyroid epithelial cell line (HTori-3) by exposure to gamma or alpha irradiation. SKY analysis revealed multiple translocations and, combined with G-banding, allowed the definition of targets for positional cloning of tumor related genes.

Fourier transform infrared fiberoptic evanescent wave (FTIR- FEW) spectra, in the middle infrared (MIR) region, of human normal skin tissue, and melanoma and benign tumors were analyzed using chemical factor analysis (CFA). The first step in CFA is to determine the rank of the factor space and in this study several model validation techniques were employed. In particular we compare results obtained from Complete Cross Validation (CCV), Binary Cross Validation (BCV), Fisher variance ratios (F-Tests), Malinowiski indicator function (IND) and significance level (%SL). All methods' results were in agreement expect for F-Tests which differed with the other methods for normal skin tissue and melanoma tumors' rank. Using the two highest ranking eigenvectors for normal skin tissue as a basis set for cluster analysis, normal skin and melanoma tumors' clusters were well separated and localized in the two dimensional factor space. The projection of benign tumors' spectral points in this factor space revealed that some of the benign tumors had already regressed into either cancerous and other abnormal skin states. Furthermore the angular disparity between normal skin tissue and melanoma tumors' eigenvectors was successfully used to discriminate between the two skin states.

The Nomarski differential interference contrast (DIC) mode is commonly used for imaging translucent biological specimens and it exhibits several major advantages over other phase contrast techniques, including a boost in high spatial frequencies in the region of focus. However, DIC images (unlike confocal) are limited by the presence of low spatial frequency blur and also by a differential shading gradient at feature boundaries, which make normal confocal visualization techniques unsuitable for feature extraction or for 3D volume rendering of focus- series datasets. To remedy this problem, we employ a neural network technique based on competitive learning, known as Kohonen's self-organizing feature map (SOFM), to perform segmentation, using a collection of statistics (know as features) defining the image. Our past investigation showed that standard features such as the localized mean and variance of pixel intensities provided reasonable extraction of objects such asmitotic chromosomes, but surface detail was only moderately resolved. In this current work, local energy is investigated as an alternative image statistic based on phase congruency in the image. This, along with different combinations of other image statistics, is applied in a SOFM, producing 3D images exhibiting vast improvement in the level of detail and clearly isolating the chromosomes from the background.

A previous paper by the authors presented an algorithm that successfully segmented organs grown in vitro from their surroundings. It was noticed that one difficulty in standard dyeing techniques for the analysis of contours in organs was due to the fact that the antigen necessary to bind with the fluorescent dye was not uniform throughout the cell borders. To address these concerns, a new fluorescent technique was utilized. A transgenic mouse line was genetically engineered utilizing the hoxb7/gfp (green fluorescent protein). Whereas the original technique (fixed and blocking) required a numerous number of noise removal filtering and sophisticated segmentation techniques, segmentation on the GFP kidney required only an adaptive binary threshold technique which yielded excellent results without the need for specific noise reduction. This is important for tracking the growth of kidney development through time.

We present a novel technique for testing of the high quality microscope objective lenses. The characterization of the lens is achieved by using a point light source approximated by a 40 nm colloidal gold bead scatterer and simultaneously measuring the field distribution in the pupil plane (pupil function) and light intensity in the image plane (point spread function). Aberrations introduced by the lens are then expanded into Zernike polynomials. The proposed technique is particularly suited for measuring apodization and vignetting effects and allows for easy measurements of the off-axis aberrations.

The simultaneous manipulation and observation of morphological features of single living cells along with the recording of functional changes related to cellular metabolism, interaction and communication in real time are of growing interest. With the advent of atomic force microscopy (AFM) structural studies of native cells became possible which in combination with adequate light microscopy give a much better resolution than light microscopy alone. However, the motion of the cell, the softness of the cell membrane and the two-dimensional growth of cells in culture limit applicability and resolution of this technique. A good mechanical fixation of living cells in a structure could be achieved by embedding cells into partially covered grooves produced in Si/SiO₂. However, for additionally optical microscopy studies a transparent material is essential. In this study we present the fabrication of different transparent three-dimensional structures in a three- layer system (Si₃N₄, SiO₂, Si₃N₄) on quartz specially sized for the trapping of living neural cells under physiological conditions. For the fabrication of the structures we utilized a combination of e-beam lithography (EBL), laser ablation, reactive ion beam etching (RIBE) and different wet etching techniques. The structures consist of a nano-net of Si₃N₄ covering a micro-cavity in SiO₂ confining enough to prevent the cell body from escaping, but not so constraining that it hinders normal growth and development. Another type of presented structures consists of cavities which are connected by covered micro-channels, hence, an observation of cell-cell interactions is also possible. Advantages of these microstructures are the trapping of cells, the stabilization of the cell membrane and the precise placement of the cells for a multitude of biological investigations.

A well-known distortion of objects in three-dimensional microscopy manifests itself as an elongation in the axial direction. Authors such as Visser and Hell have seemingly contradicted one another on the cause as well as the magnitude of the effect. We have examined these theories and performed simulations and experimental measurements to better understand the nature of the effect. We simulate point spread functions (based on the work of Gibson) taking into account the various refractive indices involved as well as the magnification, the numerical aperture, the working distance of the objective, the depth of the object under the coverslip, and the object's size. We measure the axial and lateral dimensions of digitized images of microspheres that have been 'acquired' using a simulated point spread function that changes as the depth of the object changes. These simulations are done for conventional (optical sectioning) microscopy as well as for confocal microscopy. Further, we have performed experimental measurements on real microspheres on a conventional microscope to relate theory, simulation, and practice. Our measurements and simulations show that (1) the object's size, (2) its depth under the coverslip, (3) the refractive index mismatch between the immersion fluid (n_immersion) and embedding material for the object (n_embedded), and (4) the NA of the lens play a pivotal role in the effect.

It is shown that based on spectrally selective excitation of individual molecules in the focus of a high NA lens together with position sensitive imaging sub-resolution imaging of three-dimensional structures can be realized. The feasibility of the idea is demonstrated with NA equals 0.55 optics on a model system of pentacene molecules in p-terphenal host matrix.