Implementation of one-dimensional Fourier transform techniques for the analysis of histologic sections is discussed, as is the motivation for their use. Features of the frequency domain representation derived by such a transform are shown to be related to several important diagnostic clues. The interpretation of the Fourier magnitude spectrum in histologically relevant terms is examined by means of Fourier transforms of idealized tissue simulations. Some of the perturbations of these ideal spectra produced by biologic reality are discussed. Three classic types of cervical epithelial tissue are modeled, and their representation as Fourier magnitude spectra interpreted in the light of the previous results: characteristic frequency domain signatures are obtained for each. It is concluded that these techniques may provide diagnostically important objective measures, and may be applied to otherwise intractable histologic specimens with crowded and overlapping nuclei.

The Heidelberg Polyp multiprocessor and its application to scene segmentation problems in histopathology is discussed, including ways in which the architecture can be utilized to support expert system-guided scene segmentation software, the system's current performance, and some major improvements currently being made to the system.

An expert system to guide scene segmentation in histologic sections is described. The system uses a semantic net for knowledge representation. At each node of the net, frames are associated to allow the staging of additional information. Scene segmentation is the result of model-based reasoning, supported by a digital image processing library and adaptive selection of segmentation procedures to resolve local segmentation difficulties.

The understanding of muscle contraction and relaxation requires the quantitation of movement at the sub-micron level in living cells. Two complementary non-RS-170 imaging systems used for authentic real time measurement of contractile dynamics are described and compared. Images from isolated skeletal or cardiac muscle cells are projected by an optical microscope onto single line or area charge-coupled device (CCD) photodiode arrays. These data are digitized and stored for subsequent image processing and analysis. The inherently low contrast muscle striation patterns are enhanced and their rapid movement measured with an accuracy at least an order of magnitude greater than traditional limits of optical resolution. The features of each image format are complementary and when combined provide the maximum overall information in time and space.

Scanning tunneling microscopy and atomic force microscopy, denoted here scanning tip microscopy, are two powerful novel techniques for imaging surfaces with atomic resolution. We describe the underlying principles of these two techniques with special emphasis on an instrument developed in our laboratory that uses a laser diode to detect minute deflections of a tip as it raster scans the surface of a sample. Applications of these techniques to research in biology are assessed and their relative merits discussed.

An imaging flow cytometer is described in this paper. It possesses the high cell processing rate of a flow cytometer and the high image resolution of an automated microscope. It is a flow cytometer to which has been added a cell detector and velocity measuring circuit and a microscope objective lens that focuses cell illuminated by a laser onto a two-dimensional charge-coupled device (CCD) array. Optical sensing and imaging are carried out in a water filled chamber to reduce optical abberration. The time-delay and integration (TDI) technique is utilized for image acquisition. Image processing and classification is to be carried out in real time by a n-channel, metal-oxide-semicoductor (nMOS) parallel processor array. This imaging flow cytometer can acquires image at a flow rate of 0.5 m/s and in the future will be capable of analysing and sorting cells at high speed on the basis of simultaneous morphology as well as cytochemistry and immunology.

Eigenanalysis is a powerful mathematical technique for analyzing matrices of data. With the data matrix constructed from a digitized image of a chromosome, this technique can be used to extract the features of the image, such as the chromosome banding pattern. The study of chromosome banding patterns represented by their pixel values in the images is based on eigenanalysis of the correlation or covariance matrix. Since the resulting eigenvectors are orthogonal, the information in each vector is excluded from all other vectors. Alternatively, the singular value decomposition method can be used to represent the data matrix as sum of its outer products, thereby avoiding the construction of a correlation/covariance matrix. Both procedures allow the sorting of information according to its significance, because the most significant information is associated with highest eigenvalues and corresponding eigenvectors. Consequently, the original data can be reconstituted using only the significant information. The advantage of this processing is that the preparatory artifacts and noise in the image are removed from the data before a recognition procedure is begun. An additional feature of this technique is that multiple data sets can be combined and processed simultaneously to establish, using objective statistical criteria, prototypes for each chromosome. Accumulative analysis improves the prototypes, and consequently the classification procedure. Features from prophase human chromosome number four have been to illustrate the eigenanalysis. Chromosomes from different spreads and individuals were used. Comparison of our statistically determined prototype with schematic idiotype from the literature shows significant improvement in recognition for all chromosomes, reconstituted at the level of only the most significant eigenvector. This type of analysis can be used for objective comparison of the various chromosomal banding patterns created by Giemsa, fluorescent dyes, monoclonal antibodies, and restriction enzymes. The ultimate objective is to relate various banding patterns, identified by eigenanalysis, to the genome structure.

Fluorescent probes, widely used in biochemical sciences, can be excellent tools. However, no single probe is good in all situations. For some cellular studies, quantum yield and stability might be the most important traits; whereas, for intramolecular observations, a long lifetime is an attractive fluorescent feature. Particularly useful are new fluorophores that can be excited by the argon laser at its principal 488 and 514 nm lines for uses in flow cytometry, confocal microscopy, DNA sequencing and other areas and fluorophores that have higher photostability than fluorescein. As our ability to measure the properties of the fluorophores increases and becomes more exact, we must continually reexamine the fluorophores we know and reassess the meaning of the data they supply. In addition, we must be aware of the emergence of new biologically applicable fluorescent probes. An example of one of these new species is discussed here. Since most common dyes are non- or weakly-fluorescent, fluorescent probes have been derived from a relatively limited number of aromatic fluorophores. Completely new fluorophores are rarely developed. Wories1 described synthesis and some spectral characterization of a set of dyes having fluorescein-like spectral properties from the 2,2'-pyrromethene-1,1'-borondifluoride complex. Use of similar derivatives of this new fluorophore as laser dyes has recently been reported. Kang and Haugland have prepared a variety of reactive derivatives from this fluorophore for use as fluorescent probes and for preparation of other probes. Among the probes prepared by Wories, et al. is the sodium salt of 3,3',5,5'-tetramethy1-2,2'- pyrromethene-1,1'-borondifluoride-4-sulfonate complex (here called Bodipysulfonate; structure shown in Fig. 1). We have measured the absorption, fluorescence, and lifetimes of this compound, as well as its non-sulfonated parent compound, 3,3',5,5'-tetramethy1-2,2'-pyrromethene-1,1'-borondifluoride which we have termed Bodipy (trademark of Molecular Probes, Inc.) and the complex bound to two different proteins. The results are presented here.

In Multi-Parameter Fluorescence microscopy (MPF), as many as five fluorophores have been used simultaneously, to study structural and physiological properties of cells in culturel. Each fluorophore is examined and evaluated by using a filter set (excitation filter, emission filter and dichroic reflector) specific for that fluorophore. The filters for each fluorophore cannot be chosen in isolation, but depend on what other fluorophores are to be used. Thus filter selection depends on a large number of spectra, and the task of optimization becomes complex as the number of fluorophores is increased, so that the excitation and emission wavelengths begin to overlap. This paper presents a mathematical statement of the problem, a method for computer solution, and recommendations for standard presentation of fluorescence spectra. Related efforts by other workers has approached two problems - analysing envelope spectra in terms of their known components, and finding the best quasi-monochromatic wavelengths to quantify components by ratioing. In contrast, the present work considers specific excitation of each fluorophore, considers the filter bandpasses in relationship to the detector sensitivity, and also considers simultaneously the excitation and the emission aspects of fluorescence. Though primarily directed to fluorescence microscopy, this work has application in other fields, such as flow cytometry, where multiple fluorophores are used. The mathematical statement also gives insight into the process of quantifying fluorescence.

In the last several years a number of approaches have been developed that permit detection of nucleic acid hybridization using fluorescence techniques. These have several advantages over the previously available radioactive procedures. Among these are much higher spatial resolution, the speed with which the results are available, the reduced hazard, and the possibility of distinguishing the binding of multiple probes using different fluorochromes. Exploitation of this potential requires application of optical techniques of increasing sophistication such as development of new fluorophores, improved methods of visualizing multiple dyes in the same sample, application of image processing to extract quantitative information, and optimized optical sectioning for three dimensional reconstructions.

Absorption and fluorescence spectroscopic studies are reported here for nine new fluorescent probes recently synthesized in our laboratories: four pyrene derivatives with substituents of (i) 1,3-diacetoxy-6,8-dichlorosulfonyl, (ii) 1,3-dihydroxy-6,8-disodiumsulfonate, (iii) 1,3-disodiumsulfonate, and (iv) l-ethoxy-3,6,8-trisodiumsulfonate groups, and five [7-julolidino] coumarin derivatives with substituents of (v) 3-carboxylate-4-methyl, (vi) 3- methylcarboxylate, (vii) 3-acetate-4-methyl, (viii) 3-propionate-4-methyl, and (ix) 3-sulfonate-4-methyl groups. Pyrene compounds i and ii and coumarin compounds v and vi exhibit interesting absorbance and fluorescence properties: their absorption maxima are red shifted compared to the parent compound to the blue-green region, and the band width broadens considerably. All four blue-absorbing dyes fluoresce intensely in the green region, and the two pyrene compounds emit at such long wavelengths without formation of excimers. The fluorescence properties of these compounds are quite environment-sensitive: considerable spectral shifts and fluorescence intensity changes have been observed in the pH range from 3 to 10 and in a wide variety of polar and hydrophobic solvents with vastly different dielectric constants. The high extinction and fluorescence quantum yield of these probes make them ideal fluorescent labeling reagents for proteins, antibodies, nucleic acids, and cellular organelles. The pH and hydrophobicity-dependent fluorescence changes can be utilized as optical pH and/or hydrophobicity indicators for mapping environmental difference in various cellular components in a single cell. Since all nine probes absorb in the UV, but emit at different wavelengths in the visible, these two groups of compounds offer an advantage of utilizing a single monochromatic light source (e.g., a nitrogen laser) to achieve multi-wavelength detection for flow cytometry application. As a first step to explore potential application in cancer cell diagnostics, we have found that at least two of these probes are preferentially taken up by cancerous lymphocytes as compared to normal peripheral blood lymphocytes. The feasiblity of using these probes in diagnosing malignant cells in the body fluid of cancer patients directly on a fluorocytometer is presently being investigated.

PCNA also called cyclin, is a nuclear protein and its synthesis is directly correlated to cell proliferation. At least two intracellular populations of PCNA/cyclin exist, one of which is tightly associated with DNA replication sites. This S-phase associated protein can be used as an antigenic marker for cell growth and more specifically for the S-phase. Data from clinical studies indicate that the expression of PCNA/cyclin , evaluated by immunocytochemistry or flow cytometry, can be a useful diagnostic marker for proliferating cells. In the future, the characterization of tumors might include multiparameter analysis of the simultaneous expression of multiple cell proliferation markers at different cell cycle stages.

Flow Cytometry facilities are well established and provide immunophenotyping and DNA content measurement services. The application of immunophenotyping has been primarily in monitoring therapy and in providing further information to aid in the definitive diagnosis of immunological and neoplastic disease such as: immunodeficiency disease, auto immune disease, organ transplantation, and leukemia and lymphoma. DNA content measurements have been particularly important in determining the fraction of cycling cells and presence of aneuploid cells in neoplasia. This information has been useful in the management of patients with solid tumors.

Human cancer cells selected for resistance to several structurally unrelated cytotoxic drugs are known to display plasma membrane alterations such as amplified levels of a variety of glycoproteins, modifications in lipid composition, alterations in membrane fluidity and increased cellular fragility to osmotic shock. We have studied the plasma membrane fluidity of HL60 human leukemia cells and MCF-7 human breast cancer cells that have been selected for acquired resistance against the cytocidal effects of the anthracycline anticancer drug Adriamycin. Fluidity measurements were accomplished by evaluating the fluorescence anisotropy of the plasma membrane specific probe trimethylamino-1,6-dipihenylhexatriene (TMA.DPH) bound to whole, living cells. TMA.DPH anisotropy values for MCF-7 sensitive and 12-fold resistant cells were 0.306 and 0.285, respectively, while anisotropy values for HL-60 sensitive and 80-fold resistant cells lines were 0.310 and 0.295, respectively. In all cases, cell viability exceeded 97% and anisotropy values were subject to a day-to-day uncertainty of ±2%. Our results demonstrate that increased plasma membrane fluidity apparently accompanies the development of resistance in both cell lines. Because it is known that increased membrane fluidity results in significantly decreased Adriamycin binding in artificial membrane systems, we propose here that decreased drug associations with fluidized, plasma membrane lipid bilayer regions may be a mechanism which contributes, in part, to the reduced rates of drug accumulation observed in HL60 and MCF-7 cells resistant to Adriamycin.

Abnormalities in cellular DNA content and proliferative rate are observed not only in cancer, but may be seen in premalignant disease as part of the multistep process of progression towards malignancy during carcinogenesis. Flow cytometry can be used to document these premalignant changes. Several gastrointestinal diseases provide good models for studying these abnormalities as early detectors which may select those patients at greatest risk for progression to cancer.

Flow cytometry offers an ideal new tool for evaluating aquatic bacteria because of the many analyses that can be made on each cell, sensitivity exceeding that of the light microscope, speed, simplicity of sample preparation, ability to sort, and the ideal density of bacterial populations in natural systems for direct analysis. Capabilities are demonstrated here by resolving the comparatively small bacteria in seawater samples as well as filter-fractionated subpopulations of them. Histograms from under-ice lakewater indicated even smaller and more difficult to resolve organisms; these are shown as well. Several suggestions are made for improving the effectiveness of flow cytometry for bacterial analyses. Resulting enhanced sensitivity should increase the precision with which flow cytometric measurements of aquatic bacteria can be made and improve the already substantial utility of this instrumentation as a diagnostic tool for the larger and better-known terrestrial symbiotic and pathogenic types.

The traditional domain of flow cytometry has been the measurement of properties of discrete biological particles - primarily cells or chromosomes. In actuality, most flow cytometric fluorescence measurements of cellular properties are determinations of the amount of a specific type of molecule such as DNA or a cell surface antigen. Since the early developments in flow cytometry, there has been a continuing quest for more sensitive instruments and for techniques to detect and quantitate lower levels of cellular components. The effort described here is directed at increasing the sensitivity of molecular detection by laser induced fluorescence to the point that single molecules can be both detected and identified.

A new type of analytical and preparative cytometric instrument was developed. The instrument combines image analysis and machine vision with single cell and chromosome manipulation by means of optical trapping. A proof-of-principle instrument, OCAM, has the ability to locate and analyze biological particles inside an enclosed manipulation chamber, as well as the ability to move and position particles according to preprogrammed protocols. Preliminary results and potential biological applications of such a microrobot are discussed.

Zeeman interferometry, based on the two-frequency Zeeman effect laser, is simple, stable, and has high resolution. The technique has a number of potential applications for cytometry.