A practical application of computer pattern recognition research: the Abbott ADC-500 differential classifier.

The ADC-500 is a new blood cell differential classifier manufactured by Abbott Laboratories. It performs 500-cell leukocyte differentials on both normal and abnormal cells, evaluates red cell morphology and estimates platelet sufficiency at a rate of 40 to 50 samples per hour in stand-alone operation. The ADC-500 system consists of a spinner which prepares a uniform blood monolayer on a slide, a stainer which reproducibly stains the slide with Wright's stain, an encoder which attaches an instrument and human readable identification to the slide and an analyzer which accepts a stack of up to 50 slides, evaluates these slides and prints the results and the slide identification on report forms. The system's analysis rate, which represents a 5- to 10-fold increase over other commercially available differential counters, requires a number of specialized techniques for its realization. One key to this performance is the development of a high speed X-Y slide positioning stage which can move to a new cell and settle in 50 msec. Another is the high degree of parallelism used in the system structure and the pipelining of the data processing. A third is the development of uniform and repeatable sample preparation modules. Within the analyzer module, the autofocus, white cell acquisition and high resolution cell analysis systems are independent and operate in parallel. At the same time within the high resolution cell analysis system, one cell is acquired; the digitized image of a second processed; and a third is classified using pattern recognition techniques. All of these tasks, except focus, are under the control of a minicomputer system. Tests of the system reveal good accuracy and an improvement in precision due to the increase in the number of counted cells.     

Detection of sexually transmitted diseases by urethral cytology, the ignored male counterpart of cervical cytology

The detection of sexually transmitted diseases by urethral cytology was investigated in 270 men examined by urethral swabbing smears. Each sample was used to prepare a wet mount smear and smears for staining by the Papanicolaou, Gram and methylene blue techniques. A fifth smear was used for direct staining with fluorescein-conjugated monoclonal antibodies for the detection of Chlamydia trachomatis. The smears were examined for cytoplasmic and nuclear changes as well as for pathogenic organisms and inflammatory changes. Infections with Chlamydia trachomatis, Neisseria gonorrhoeae and human papillomavirus (HPV) produced distinctive cytologic patterns similar to those seen in cervicovaginal smears from women. The patterns in candidiasis, trichomoniasis and herpes simplex virus infection were not as diagnostic. Particularly noteworthy were the nuclear alterations, which appeared to be proplastic in HPV infection but retroplastic in Chlamydia infection. The results of this study indicate that urethral cytology would be an invaluable addition in diagnosing sexually transmitted diseases in men, particularly in the case of Chlamydia and HPV infections. The monomorphic structure of urethral columnar epithelium, as compared to the cervical epithelium, seems to result in a clearer and more constant response to pathogenic infections, as seen in the resulting smears.     

Variability of erythrocyte size and hemoglobin content observed in man and four selected mammals

A new measurement technique employing light scattering at different angles has been developed for analysis of blood cells. This method which is part of the Technicon H*1 system, a hematology analyzer designed for routine processing of human blood samples, allows the independent measurement of cell volume and hemoglobin content of isovolumetrically sphered red blood cells. Analysis with this instrument of the blood of humans, dogs, rabbits, rats and mice demonstrates that in addition to the expected differences in hematologic parameters, the intrasample distribution of cell hemoglobin is species dependent. In general, cell hemoglobin content is more tightly controlled for the other mammals when compared to humans. In particular, the dogs tested showed the least variability in cell hemoglobin content both within species and within sample.     

An immunogold-silver staining method for detection of cell surface antigens in cell smears

We developed an indirect immunogold-silver staining method for detection of leukocyte cell surface antigens in cell smears. Air-dried and fixed cytocentrifuge preparations or smears of peripheral blood leukocytes were incubated with monoclonal antibodies (MAb) and colloidal gold-labeled secondary antibodies. The preparations were post-fixed and silver enhancement was performed. The smears were counterstained with May-Grunwald-Giemsa and examined in brightfield light microscopy. The morphology of the cells was well preserved. Leukocytes reacting with the MAb showed black granules on their surface membranes. The intense immunostaining and the low background allowed a rapid enumeration of the positive cells. The labeling could be detected with high sensitivity by epipolarization microscopy. This immunogold-silver staining method was used to quantify T- and B-lymphocytes and natural killer cells in buffy coat smears of normal adult blood. These lymphocyte subsets correlated well with those obtained in smears with the alkaline phosphatase-anti-alkaline phosphatase (APAAP) method and with those found by labeling of mononuclear cells in suspension with immunogold-silver staining. This immunogold-silver staining method forms a good alternative to immunoenzyme methods for study of hematologic cells. In addition, it could be a general procedure for detection of cell surface antigens in all kinds of cell smears.     

Measurement of erythrocytes on diagnostic slides by scanning electron microscopy

OBJECTIVE: To determine whether the measured sizes of erythrocytes in both paraffin-embedded sections and air-dried blood smears differ from values published in standard texts. STUDY DESIGN: Routinely prepared surgical pathology slides as well as an air-dried blood smear were viewed with a scanning electron microscope. Erythrocytes were measured using the instrument software. RESULTS: Erythrocyte size in the peripheral blood smear correlated well with textbook values, 7.2-7.9 microns. However, red blood cells within sectioned material from several laboratories showed a prominent decrease, ranging from 25% to 35%, as compared to textbook values, about 7 microns. CONCLUSION: Since cytologists and surgical pathologists often use the erythrocyte as a convenient marker on diagnostic slides, attention should be given to these observations in making sizing judgments.     

Measurement of DNA content in single cells morphologically identified on smears

A method was developed for measuring the nuclear DNA content in single cells previously identified on a bone marrow smear stained by the Wright-Giemsa method. The smear was first photographed and the location of individual cells, identified by morphology, was recorded on a cell map. The smear was then bleached with 50% acid ethanol and absolute methanol, and re-stained by the Feulgen method in 0.05% pararosaniline Schiff's reagent (pH 2.3) at 7 degrees C for 10 min. Nuclear red fluorescence was observed and the intensity of this fluorescence was proportional to the amount of DNA after prior irradiation of smears with green light for 9 hr. The method is useful for measuring cell DNA content in heterogeneous cell populations when morphological cell identification is required.     

Interpretation of atypical patterns encountered when using a flow cytometry-based method to detect residual leukocytes in leukoreduced red blood cell components

BACKGROUND: Universal leukoreduction of blood components is becoming the standard of care. Flow cytometry methods are being used for quality control of the leukoreduction process. METHODS: We provide an atlas of atypical flow cytograms generated by a commercial LeucoCOUNT assay that was used to enumerate residual leukocytes in leukoreduced red blood cell components. Numeric results are derived from a flow cytogram generated by the assay. RESULTS: Three types of atypical flow cytogram patterns were observed during process validation or routine quality control of leukoreduced red blood cell components. (a) Fixation artifact: Fixation of control or test samples can alter the staining intensity compared with fresh cells. (b) "Rain" pattern: Flow cytometry methods count slightly damaged leukocytes not removed during leukoreduction. Slightly damaged leukocytes appear on a flow cytogram like "rain" falling from a well-defined "cloud" of intact residual leukocytes. Discrepancies between automated flow cytometry results and subjective manual counting methods can occur. (c) Autofluorescence-debris pattern: Cell debris and age-related changes in the sample can cause shifts in the fluorescence staining pattern, resulting in erroneous test results. CONCLUSION: Review of flow cytograms is essential for accurate reporting of flow cytometry-based methods for enumerating residual leukocytes in leukoreduced blood components.     

Rapid detection of malaria and other bloodstream parasites by fluorescence microscopy with 4'6 diamidino-2-phenylindole (DAPI).

DAPI is a fluorescent dye which appears to complex specifically with DNA. We have used this probe to detect and identify malarial infections by fluorescence microscopy. Experiments were conducted using Plasmodium berghei yoeli--infected mouse blood, P. lophurae--infected duck blood, and P. vivax--infected human blood. Infected avian blood was used to detect parasites within nucleated erythrocytes. Control blood smears from uninfected hosts revealed fluorescence only in the leukocytes of mammalian blood or in nuclei of leukocytes and erythrocytes of avian blood. Cytoplasmic staining of red blood cells was absent in all controls. In contrast, the cytoplasm of infected red blood cells was stippled with fluorescence centers. Ring forms, trophozoites, segmenters, and merozoites frequently were observed. This simple procedure can be applied directly to routine clinical analysis, as well as experimental procedures, DAPI can also be used to stain other parasites, including nuclei in microfilariae.     

Artificial intelligence in automated classification of rat vaginal smear cells.

Microscopic examination of vaginal smears has been used routinely to determine the stage of the estrous cycle of female rats in reproductive research. The stage of the estrous cycle is based on relative counts of nucleated epithelial cells, cornified epithelial cells and leukocytes. The purpose of this project was to explore automation of vaginal smear analysis using image processing and artificial intelligence techniques. A fully connected back-propagation neural network was used to locate all potential objects in a digitized scene. A unique algorithm was then employed to center a subsequent sampling box to collect pixel intensity values from the red and green components of each image. A final neural network was used in the classification of cell type. Neural networks were used because of their ability to generalize among input patterns and to tolerate extraneous noise due to variations in staining artifacts and aberrant illumination of the microscope field. This preliminary cell diagnosing system not only provides the basis for the fully automated system but also provides a method by which many other cytologic image processing problems can be automated.     

Preliminary observation of blood cells and hematopoietic organs in Lutjanus herenbergi and Lutjanus flaviflammus

The morphology of differentiated and differentiating cells of the red and white series in Lutjanus herenbergi and in Lutjanus flaviflammus is described. Early stages of red and white blood cells may be found only in smears of hemopoietic organs. Polychromatic erythroblasts, myelocytes and lymphoblasts may also occasionally be found in blood smears. Mature blood cells may be found both in blood smears and in hemopoietic organs. Differential white cell counts seem to demonstrate that the granulocytic series elements are the most common leukocytes in blood smears. Almost all granulocytes may be classified in the first three Arneth classes. An analysis of hemopoietic organs in these species was also performed. It was found that the only organs carrying on a hemopoietic function are the kidney and the spleen. The kidney is essentially a site of granulocytic differentiation while the spleen is a lymphopoietic organ. An erythropoietic activity may generally be observed in the kidney although weak erythropoietic activity may at times be found in the spleen.     

Four-parameter white blood cell differential counting based on light scattering measurements

Measurement of the depolarized orthogonal light scattering in flow cytometry enables one to discriminate human eosinophilic granulocytes from neutrophilic granulocytes. We use this method to perform a four-parameter differential white blood cell analysis. A simple flow cytometer was built equipped with a 5-mW helium neon laser that measures simultaneously four light scattering parameters. Lymphocytes, monocytes, and granulocytes were identified by simultaneously measuring the light scattering intensity at angles between 1.0 degrees and 2.6 degrees and angles between 3.0 degrees and 11.0 degrees. Eosinophilic granulocytes were distinguished from neutrophilic granulocytes by simultaneous measurement of the orthogonal and depolarized orthogonal light scattering. Comparison of a white blood cell differentiation of 45 donors obtained by the Technicon H-6000 and our instrument revealed good correlations. The correlation coefficients (r2) found were: 0.99 for lymphocytes, 0.76 for monocytes, 0.99 for neutrophilic granulocytes, and 0.98 for eosinophilic granulocytes. The results demonstrate that reliable white blood cell differentiation of the four most clinically relevant leukocytes can be obtained by measurement of light scattering properties of unstained leukocytes.     

Continuous measurement and analysis of staining kinetics by flow cytometry

The measurement of color development with time in cells following the start of a staining reaction is of interest in a number of biological systems. These include the subsets of peripheral white blood cells after acridine orange staining, the uptake by cells and nuclei of fluorescent agents, especially antitumor drugs, and measurement of intracellular enzyme kinetics using fluorogenic or absorbing substrates. The present work describes a simple computer program for analyzing flow cytometric (FCM) data versus time, including both the population kinetics of color development and the variability of staining speed within one population of cells. A single-channel absorption measurement in flow (Technicon Hemalog D) was used to record peroxidase kinetics in peripheral blood cells. Every 5 s, a 64-channel absorption histogram was recorded, up to a maximum of 64 histograms. The data were then analyzed by a computer program which searched for the peak channel of each histogram. A least-squares fit was computed for these maxima. The asymmetries of the 64 absorption histograms were compared to see if there was more than one population present with different time constants. Although developed for enzyme kinetic measurements, this program may have wider usefulness in any measurements of time-dependent phenomena by FCM.     

Simultaneous detection of T-, B- and D-rosette forming lymphocytes and null cells in humans]

For the identification of T-, B-, and D-rosette-forming lymphocytes and the "null" cells in the human peripheral blood a simultaneous reaction of rosette-forming cells with the use of zymosan-complement, sheep red blood cells, fixation and staining of smears in extraction of the lymphocytes tested with the aid of verographin may be recommended.     

The value of multiple preparations in the diagnosis of malignant pleural effusions. A cost-benefit analysis

A cost-benefit analysis of five techniques employed in processing 108 malignant pleural effusions for cytopathologic examination was performed. Ethanol-fixed, Papanicolaou-stained smears were positive in 68% of the effusions, air-dried Diff-Quik-stained smears in 66%, Cytospins in 83%, cell blocks in 85% and Millipore filters in 85%. Examination of one air-dried smear and one ethanol-fixed smear yielded a diagnostic sensitivity of 82%. Using a combination of two smears and one of three concentrating techniques (Cytospin, cell block or MIllipore filter) would have provided a diagnosis in over 90% of the effusions. The use of four or more preparations provided more sensitivity than did three preparations and decreased the likelihood that a malignant diagnosis would be based on the findings in only one preparation. The costs related to disposable materials and the College of American Pathologists (CAP) work load estimates for specimen preparation, staining and screening were compared. These data may be useful in developing optimal protocols for pleural effusion preparation in laboratories with specific work load requirements and limited resources.     

Concentrating and Staining Bone Marrow; An On-the-Slide Technique

A 0.5-1 ml sample of bone marrow is aspirated into a syringe containing 3 drops of 15% K₂-EDTA and an additional 1-2 drops of the EDTA solution previously placed on a slide, is then drawn into the syringe. All of the contents are ejected onto this slide, which is carefully tilted 2 or 3 times to an angle of 5-10°, and the edge brought to the center of another slide. The slide with the aspirate is then slowly tilted to 80-90°. Most of the blood and part of the marrow will drain off, leaving spicules of marrow and some blood on the original slide. A small drop of this concentrated marrow is dragged off with the edge of a third slide and deposited about 2 cm from the edge of a fourth slide on which the smear is to be made. The smear is made by bringing a clean (smearing) slide to the slide with the deposited marrow with flat surfaces parallel and the edges at a 90° angle. With gentle pressure, the smearing slide is pushed toward the empty end of the slide upon which the smear is made. This separates the marrow from the circulating blood. Before staining the smear is air dried and heated in an oven at 120-125 C for 2 min; or alternately for satisfactory but less uniform results the smear is heated over a microburner for 10 sec; then the smear is covered with 1 part of undiluted Wright's stain for 30—45 sec which is then diluted with 2 parts of a solution of 0.1-0.2 gm of Na₂S₂O₃ in 1 liter of distilled water and stained for 10-13 min with this diluted stain. Smears made in this manner have 3 concentric zones; the central zone contains the myeloid tissue; the middle, erythropoetic tissue; the outer, a mixture of blood and marrow.     

LeukocyteMask: An automated localization and segmentation method for leukocyte in blood smear images using deep neural networks

Digital pathology and microscope image analysis is widely used in comprehensive studies of cell morphology. Identification and analysis of leukocytes in blood smear images, acquired from bright field microscope, are vital for diagnosing many diseases such as hepatitis, leukaemia and acquired immune deficiency syndrome (AIDS). The major challenge for robust and accurate identification and segmentation of leukocyte in blood smear images lays in the large variations of cell appearance such as size, colour and shape of cells, the adhesion between leukocytes (white blood cells, WBCs) and erythrocytes (red blood cells, RBCs), and the emergence of substantial dyeing impurities in blood smear images. In this paper, an end‐to‐end leukocyte localization and segmentation method is proposed, named LeukocyteMask, in which pixel‐level prior information is utilized for supervisor training of a deep convolutional neural network, which is then employed to locate the region of interests (ROI) of leukocyte, and finally segmentation mask of leukocyte is obtained based on the extracted ROI by forward propagation of the network. Experimental results validate the effectiveness of the propose method and both the quantitative and qualitative comparisons with existing methods indicate that LeukocyteMask achieves a state‐of‐the‐art performance for the segmentation of leukocyte in terms of robustness and accuracy .     

Removal of blood components from cervical smears: implications for cancer diagnosis using FTIR spectroscopy

Red and white cell lysis buffers were investigated to determine their ability to remove blood components from cervical smears prior to IR microspectroscopy. A white cell lysis buffer was effective in increasing the spectral reproducibility and sample homogeneity and in reducing the presence of inflammatory exudate, particularly leukocytes, from cervical smears. The reduction of leukocytes appeared to cause abnormal samples to be grouped with normal samples, resulting in poor discrimination. Despite differences between the cytological and histological diagnoses of cervical abnormalities, the results indicate that the differences seen in the spectra of exfoliated malignant and nonmalignant cervical cells could be due to the presence of inflammation.     

Fluorescent antibody localization of myosin in the cytoplasm, cleavage furrow, and mitotic spindle of human cells

We have studied the distribution of myosin molecules in human cells using myosin-specific antibody coupled with fluorescent dyes. Rabbits were immunized with platelet myosin or myosin rod. They produced antisera which precipitated only myosin among all the components in crude platelet extracts. From these antisera we isolated immunoglobulin-G (IgG) and conjugated it with tetramethylrhodamine or fluorescein. We separated IgG with 2-5 fluorochromes per molecule from both under- and over-conjugated IgG by ion exchange chromatography and used it to stain acetone-treated cells. The following controls established the specificity of the staining patterns: (a) staining with labeled preimmune IgG; (b) staining with labeled immune IgG adsorbed with purified myosin; (c) staining with labeled immune IgG mixed with either unlabeled preimmune or immune serum; and (d) staining with labeled antibody purified by affinity chromatography. In blood smears, only the cytoplasm of platelets and leukocytes stained. In spread Enson and HeLa cells, stress fibers stained strongly in closely spaced 0.5 mum spots. The cytoplasm stained uniformly in those cells presumed to be motile before acetone treatment. In dividing HeLa cells there was a high concentration of myosin-specific staining in the vicinity of the contractole ring and in the mitotic spindle, especially the region between the chromosomes and the poles. We detected no staining of erythrocytes, or nuclei of leukocytes and cultured cells, or the surface of platelets and cultured cells.     

Detection of S-Phase Cells in Smear Cytology Using in Vitro Bromodeoxyuridine Labeling

A technique Is described for rapid detection of S-pha?e cells of tumor tissues in smear specimens using bromodeoxyuridine (BrdU) immunostaining. Mouse NR-S1 tumors and human tumor specimens were prepared for smear cytology after incubation in RPMI 1640 culture medium containing 200 μM BrdU at 37 °C under 3 atm for 1 hr. Samples were fixed in 70% ethanol for 30 min and used immediately or air dried for 30 min. Samples were then denatured in either 4 N HC1 or 0.07 N NaOH to prepare partially single-stranded DMA. Fixation with air drying for 30 min followed by 30 min in 70% ethanol and 1 min denaturation with 0.07 N NaOH resulted in satisfactory staining quality. Cultured tumor specimens were processed for routine paraffin sections after smears were made for cytology. The labeling indices of the smear specimens and of the paraffin sections gave similar results. This technique should be useful in evaluating the cell proliferative potential of tumor tissue in smear cytology without processing paraffin sections.     

Parasitemia in PCR‐detected Plasmodium and Haemoproteus infections in birds

Most comparative studies of avian blood parasites based on visual inspection of smears have reported Haemoproteus infections to be more prevalent than Plasmodium infections in both tropical and temperate locations. Recently, molecular techniques have increased our ability to detect infections often missed on blood smears. Here we quantify the bias in prevalence resulting from unrecognized infections by examining blood smears of infected passerine birds from the West Indies (312 individuals) and the Ozark Mountains of southern Missouri (134 individuals) for which we could identify parasites based on cytochrome b sequences. In the West Indian sample, 63 of 179 Haemoproteus infections (35%) and 121 of 133 Plasmodium infections (91%) were not detected among ca. 2,800 red blood cells examined per smear. In the Missouri sample, 19 of 77 Haemoproteus infections (25%) and 31 of 57 Plasmodium infections (54%) were not detected among ca. 10,000 red blood cells examined. Clearly, visual inspection of blood smears at this level of effort fails to recognize many malaria parasite infections ascertained by PCR screening, and this bias for Plasmodium parasites exceeds that for Haemoproteus parasites. The lower prevalence of Plasmodium compared to Haemoproteus reported in comparative studies based on blood smears likely reflects differences in detection rather than infection rates. Estimates obtained from visual inspection of blood smears would appear to be more indicative of parasite virulence and how well host individuals control infections than of the prevalence of infections in host populations.