A very simple OCT system has been developed, based on a Linnik interferometric microscope with its reference mirror mounted on a piezoelectric translator. The geometrical extension of the optics allows efficient illumination of this device with a low power (3 W) light bulb, yielding full field interferometric images at 50 Hz acquisition rate with a fast CCD camera. Due to the very broad spectral width of the light source and camera response, a longitudinal resolution of 1.5 micrometers is achieved. Tomographic images of cell smears are shown.
En-face optical coherence tomography (OCT) employing parallel heterodyne detection technique is demonstrated for three-dimensional microscopy. To enable the use of commercially available CCD cameras as two-dimensional heterodyne detector arrays in OCT imaging, a frequency synchronous detection method is employed. Depth-resolved, full-field images are acquired at 30 Hz video-rate without lateral scanning.
In confocal microscopy manifold combinations of dyes are used. The entire wavelength range from near-ultraviolet to near-infrared is used to excite these dyes. Their emission is then detected primarily in the visible range. In addition to the number of dyes, the number of possible excitation laser lines is already large and constantly increasing. Due to the almost unlimited number of possible combinations, programmable devices are required. This is true for excitation-modules, beam-splitters and detection modules. Systems using filters with fixed spectral properties can practically fulfill the requirements for only a very small subset of applications. Programmable devices have been realized for excitation using Acousto Optical Tunable Filters (AOTF's). Freely definable spectral detectors are available for a perfect adaptation to the desired emission bands. The missing link for a completely filter-free design is now introduced using an Acousto Optical Beam Splitter (AOBS).
The use of an annular pupil plane filter may be used to increase the depth of focus of an objective lens without significant deterioration of the lateral resolution. However this approach is very inefficient since most of the illumination light is blocked by the annular filter. We describe a method which uses a diffractive optical element to increase significantly the depth of focus but with dramatically increased light efficiency.
The direct-view microscope is a confocal microscope which allows faster image acquisition rates than typical confocal scanning optical microscopes through the use of a pinhole array rather than the usual single pinhole. We present a theoretical investigation of the effects of source coherence on optical sectioning in direct-view microscopy. As a first step we present an equation which describes the optical sectioning strength of a coherent source brightfield DVM employing an infinite pinhole array. By simulation of both this and the finite array equation which show the existence of certain `principal' sidelobes which are likely to represent the most problematic artifact of coherent source imaging. By further analysis of the infinite array equation, we arrive at an expression which describes the defocus positions where the principal sidelobes occur. Finally, we move on to show how rectangular arrays are predicted, by the infinite array equation, to outperform square arrays and we show examples of this.
Multidimensional fluorescence microscopy data are often collected under imaging conditions that cause aberrations. For proper deconvolution of such data, one needs to generate a Point Spread Function (PSF) that includes the aberrations. Even if it is possible to model the imaging aberrations theoretically, there remains the problem of modifying the PSF accordingly. Measured PSFs are difficult to modify, and theoretical PSFs lack lens- and system-specific features. Phase retrieval approaches can be used to produce a compact and modifiable description of a microscope system, including its measured, system-specific performance. These pupil functions can be used in a depth-dependent deconvolution to correct for geometric distortions and improve restoration of relative intensities compared to spatially invariant deconvolution using unaberrated PSFs.
Quantitative analysis of spatial and temporal concurrent responses of multiple markers in 3-dimensional cell cultures is hampered by the routine mode of sequential image acquisition, measurement and analysis of specific targets. A system was developed for detailed analysis of multi-dimensional, time-sequence responses and in order to relate features in novel and meaningful ways that will further our understanding of basic biology. Optical sectioning of the 3-dimensional structures is achieved with structured light illumination using the Wilson grating as described by Lanni. The automated microscopy system can image multicellular structures and track dynamic events, and is equipped for simultaneous/ sequential imaging of multiple fluorescent markers. Computer-controlled perfusion of external stimuli into the culture system allows (i) real-time observations of multiple cellular responses and (ii) automatic and intelligent adjustment of experimental parameters. This creates a feedback loop in real-time that directs desired responses in a given experiment. On-line image analysis routines provide cell-by-cell measurement results through segmentation and feature extraction (i.e. intensity, localization, etc.), and quantitation of meta-features such as dynamic responses of cells or correlations between different cells. Off-line image and data analysis is used to derive models of the processes involved, which will deepen the understanding of the basic biology.
Recent evidence supports the notion that biological functions of extracellular matrix (ECM) are highly correlated to its structure. Understanding this fibrous structure is very crucial in tissue engineering to develop the next generation of biomaterials for restoration of tissues and organs. In this paper, we integrate confocal microscopy imaging and image-processing techniques to analyze the structural properties of ECM. We describe a 2D fiber middle-line tracing algorithm and apply it via Euclidean distance maps (EDM) to extract accurate fibrous structure information, such as fiber diameter, length, orientation, and density, from single slices. Based on a 2D tracing algorithm, we extend our analysis to 3D tracing via Euclidean distance maps to extract 3D fibrous structure information. We use computer simulation to construct the 3D fibrous structure which is subsequently used to test our tracing algorithms. After further image processing, these models are then applied to a variety of ECM constructions from which results of 2D and 3D traces are statistically analyzed.
The development of an animal embryo is orchestrated by a network of genetically determined, temporal and spatial gene expression patterns that determine the animals final form. To understand such networks, we are developing novel quantitative optical imaging techniques to map gene expression levels at cellular and sub-cellular resolution within pregastrula Drosophila. Embryos at different stages of development are labeled for total DNA and specific gene products using different fluorophors and imaged in 3D with confocal microscopy. Innovative steps have been made which allow the DNA-image to be automatically segmented to produce a morphological mask of the individual nuclear boundaries. For each stage of development an average morphology is chosen to which images from different embryo are compared. The morphological mask is then used to quantify gene-product on a per nuclei basis. What results is an atlas of the relative amount of the specific gene product expressed within the nucleus of every cell in the embryo at the various stages of development. We are creating a quantitative database of transcription factor and target gene expression patterns in wild-type and factor mutant embryos with single cell resolution. Our goal is to uncover the rules determining how patterns of gene expression are generated.
This paper describes a set-up for high-resolution imaging with a conventional microscope that allows producing a 3D-image set from which a 3D model can be derived. The 3D-image set, is de facto, a gray value voxel model. In order to obtain such 3D image set a good administration needs to be maintained at acquisition and moreover, the acquisition must be realized in two steps. High-resolution images are built of image tiles and sophisticated algorithms are required to build a coherent image from the tiles.
We present a new method of fluorescence imaging, which yields nm-scale axial height determination and ~15 nm axial resolution. The method uses the unique spectral signature of the fluorescent emission intensity well above a reflecting surface to determine vertical position unambiguously. We have demonstrated axial height determination with nm sensitivity by resolving the height difference of fluorescein directly on the surface or on top of streptavidin. While different positions of fluorophores of different color are determined independently with nm precision, resolving the position of two fluorophores of the same color is a more convoluted problem due to the finite spectral emission widow of the fluorophores. Hence, for physically close (<λ/2) fluorophores, it is necessary to collect multiple spectra by independently scanning an excitation standing wave in order to deconvolute the contribution to the spectral pattern from different heights. Moving the excitation standing wave successively enhances or suppresses excitation from different parts of the height distribution, changing the spectral content. This way two fluorophores of the same color can be resolved to better than 20 nm. Design aspects of the dielectric stack for independent excitation wave scanning and limits of deconvolution for an arbitrary height distribution will be discussed.
Deconvolution of confocal fluorescence images by maximum likelihood estimation (MLE) was investigated for its ability to increase the information content in the images. 3-D MLE algorithms, applied to confocal image stacks, increase lateral and axial resolution and result in a finer optical section. Contrast, especially at edges, is enhanced, improving the documentation quality of high magnification images beyond that possible by confocal microscopy alone. Axial smear associated with spherical aberration was not removed by deconvolution and a lateral thinning artifact was introduced. Single confocal images can be rapidly deconvolved by 2-D MLE by applying a two-dimensional point spread function and treating them as images of planar objects. The 2-D algorithm can also deconvolve a maximum projection of a stack. The method works best when there is a minimal overlap of fluorescent structures. The projection is treated as a two dimensional object. Intensify information excluded by the projection operation cannot be recovered. Deconvolution of images acquired with the pinhole opened to increase sensitivity closely matches images acquired through an optimal opening, although in 2-D MLE, colocalization cannot be distinguished from overlap and the integrity of quantitative data cannot be guaranteed. Properly applied, MLE deconvolution increases the useful information content of confocal images.
We propose and demonstrate a method employing ferroelectric monomolecular layers, by which it is possible to precisely measure the planar light field polarization in the focus of a lens. This method allowed us to establish for the first time to our knowledge, the perpendicularly oriented field that is anticipated at high apertures. For a numerical aperture 1.4 oil immersion lens illuminated with linearly polarized plane waves, the integral of the modulus square of the perpendicular component amounts to (1.51r0.2) % of that of the initial polarization. It is experimentally proven that depolarization decreases with decreasing aperture angle and increases when using annular apertures. Annuli formed by a central obstruction with a diameter of 89 % of that of the entrance pupil raise the integral to 5.5 %. This compares well with the value of 5.8% predicted by electromagnetic focusing theory; however, the depolarization is also due to imperfections connected with focusing by refraction. Besides fluorescence microscopy and single molecule spectroscopy, the measured intensity of the depolarized component in the focal plane is relevant to all forms of light spectroscopy combining strong focusing with polarization analysis.
A scanning optical far and near-field microscopes resolution is limited by focused light spot size used to scan a sample. As a focusing element in scanning microscopes are ordinary used lenses with dimensions of rather more than a light wavelength. The focus spot size of such lens is limited by well-known diffraction limit. However, if a lens dimension is comparable to a light wavelength the Fresnel's diffraction formula of focus spot size calculation is not suitable. At present, there are no data about theoretical or experimental studies of the small lenses and even about the ones fabrication methods. This paper is considers the fabrication method of small spherical glass lenses with diameters from 1 um to 100 um and discusses some theoretical approaches to the focus spot size calculation and it experimental measurement. The lenses fabrication method is based on melting of fine glass powder with oxygen-acetylene burner with subsequent lenses selection under conventional light microscope. To calculate a spot size of the small lens the Mie theory of light scattering in near field by small spherical particles has been used. The light intensity diagrams of sphere lens of various diameters calculated by developed program are presented.
Numerical simulation of image formation in near field optical microscopy is needed to understand a relationship between near field images and actual structure of sample since the image can be differs strongly from real structure. In order to estimate the near field image formation, two different approaches are used, namely numerical solution of integral equations with boundary element method or moment method and analytical solution of light scattering problem obtained by assumption of some hypotheses. The first approach results in reasonably complicated algorithm of calculation and often causes a bed conditioned system of linear equations to be formed. The approach based on an analytical solution of scattering problem is more preferable since allow to reduce the algorithm and increase the calculations speed. A new program of simulation of near field image formation for arbitrary sample shape has been developed. The program is based on the results of analytical calculation of near field light diffraction on sinusoidal diffraction grating of arbitrary corrugation amplitude. The sample shape is presented as a number of sinusoidal gratings of various amplitudes by Fourier transformation, for each of the gratings the light scattering amplitude is calculated and the amplitudes are summarized.
The article presents a new design of a polarization microscope with a scanning liquid crystal aperture. The scanning device is based on the earlier reported Pol-Scope technique and includes a liquid crystal universal compensator. It is mounted in the front focal plane of the high numerical aperture condenser lens on the microsccope. By occluding different portions of the aperture, an oblique beam of variable tilt angle and azimuth is created for illuminating the specimen. Birefringence measurements are recorded for different mask configurations and results are evaluated to determine the retardance magnitude, azimuth and direction of optic axis of the specimen. We report measurements using small calcite crystals that confirm our theoretical predictions.