Manual Cell Counting Research Articles

Validation of an automated cell counting method for cGMP manufacturing of human induced pluripotent stem cells

Human induced pluripotent stem cells (hiPSCs) must be manufactured as advanced therapy medicinal products (ATMPs) for innovative tissue replacement clinical applications. Yet, production of hiPSCs under current Good Manufacturing Practice (cGMP) presents many hurdles, such as the large-scale cell expansion needed to reach therapeutically-relevant hiPSC doses. For the monitoring of this phase, a fast and reliable cell counting method should be used. Conventional manual cell counting by the hemocytometer method is dependent on the operator's expertise and is time-consuming. Therefore, automation of sample preparation and analysis is needed to improve precision and rapidity of hiPSC cell counting. We investigated whether an automated cell counting method could be validated for use with hiPSCs, in comparison with a reference cell counting method included in the European Pharmacopeia, 10th edition. The proposed method was the fluorescence imaging-based NucleoCounter NC-100 system, whereas the reference method was manual cell counting using a Bürker hemocytometer. The validation strategy complied with EudraLex cGMP regulations for ATMP manufacturing and ICH Q2(R1) indications for validation of analytical methods. The use of the NucleoCounter NC-100 system for automated cell counting was validated, focusing on accuracy, specificity, intra- and inter-operator reproducibility, range and linearity, showing higher precision than the manual method. The automated method can be used more effectively than the manual one for hiPSC cell counting. Thus, this piece of work paves the way for all cGMP facilities that want to pursue hiPSC manufacturing for clinical use.

Neoplastic Cellularity Assessment in Molecular Testing.

Neoplastic cellularity assessment has become an essential component of molecular oncology testing; however, there are currently no best practice recommendations or guidelines for this potentially variable step in the testing process. To describe the domestic and international practices of neoplastic cellularity assessment and to determine how variations in laboratory practices affect neoplastic cellularity assessment accuracy. Data were derived from 57 US and international laboratories that participated in the 2019 College of American Pathologists Neoplastic Cellularity Proficiency Testing Survey (NEO-B 2019). NEO-B 2019 included 29 laboratory practice questions and 5 images exhibiting challenging histologic features. Participants assessed the neoplastic cellularity of hematoxylin-eosin-stained digital images, and results were compared to a criterion standard derived from a manual cell count. The survey responses showed variations in the laboratory practices for the assessment of neoplastic cellularity, including the definition of neoplastic cellularity, assessment methodology, counting practices, and quality assurance practices. In some instances, variation in laboratory practice affected neoplastic cellularity assessment performance. The results highlight the need for a consensus definition and improved standardization of the assessment of neoplastic cellularity. We put forth an initial set of best practice recommendations to begin the process of standardizing neoplastic cellularity assessment.

Quantitative Assessment of Epithelial Proliferation in Rat Mammary Gland Using Artificial Intelligence Independent of Choice of Proliferation Marker.

Epithelial proliferation in the rat mammary gland is recommended in regulatory guidelines as an endpoint for assessment of the in vivo carcinogenic potential of insulin analogues. Epithelial proliferation is traditionally assessed by immunohistochemical staining of a proliferation marker, for example, 5-bromo-2'-deoxyuridine (BrdU) or Ki67, followed by labor-intensive manual counting of positive and negative cells. The aim of this study was to develop and validate an approach for image analysis based on artificial intelligence, which can be used for quantification of proliferation in rat mammary gland, independent of the choice of proliferation marker. Furthermore, the aim was to compare the markers BrdU, Ki67, and phosphorylated histone H3 (PHH3). A sequence of image analysis applications were developed, which allowed for quantification of proliferative activity in the mammary gland epithelium. These endpoints agreed well with manually counted labeling indices, with correlation coefficients in the range ≈0.92-0.93. In addition, all three proliferation markers were significantly correlated and could detect the variation in epithelial proliferation during the estrous cycle. In conclusion, image analysis can be used to quantify epithelial proliferation in the rat mammary gland and thereby replace time-consuming manual counting. Furthermore, BrdU, Ki67, and PHH3 can be used interchangeably to assess proliferation.

Comparison of manual and automatic cell count methods for synovial fluid: A prospective study

Background/Aim: Cell count measurement methods in the synovial fluid are still a current problem in orthopedic practice. Knowing the cell count in the synovial fluid is important for the assessment of a variety of orthopedic and rheumatologic diseases. We aimed to assess the correlation of WBC and RBC results obtained with a complete blood count analyzer with that obtained by a manual cell count. Methods: The WBC and RBC count in the synovial fluid of 43 patients undergoing gonarthrosis surgery were determined by a Mindray BC-6800 hematology analyzer. The study groups were defined as manual cell count (MC), hemogram mode (HM), and body fluid mode (BFM). All samples were analyzed twice consecutively, and the mean results were calculated. Cell counting was performed using different methods in the same samples and compared statistically. Results: The mean age of the patients was 60.9 years, and there were 17 males (39.5%) and 26 females (60.5%). The WBC and RBC counts in the synovial fluid samples were determined using manual cell count, and the HM and BFM on a Mindray BC-6800 automatic hematology analyzer. WBC counts significantly differed between MC and BFM, and RBC counts significantly differed between HM-MC and HM-BFM (P=0.001, P=0.001, P=0.001, respectively). There was a significant positive correlation between BFM and MC in WBC counts (r=0.633, P

Selective Hole Filling of Red Blood Cells for Improved Marker-Controlled Watershed Segmentation.

Manual counting and evaluation of red blood cells with the presence of malaria parasites is a tiresome, time-consuming process that can be altered by environmental conditions and human error. Many algorithms were presented to segment red blood cells for subsequent parasitemia evaluation by machine learning algorithms. However, the segmentation of overlapping red blood cells always has been a challenge. Marker-controlled watershed segmentation is one of the methods that was implemented to separate overlapping red blood cells. However, a high number of overlapped red blood cells were still an issue. We propose a novel approach to improve the segmentation efficiency of marker-controlled watershed segmentation. Local minimum histogram background segmentation with a selective hole filling algorithm was introduced to improve segmentation efficiency of marker-controlled watershed segmentation on a high number of overlapping red blood cells. The local minimum was selected on the smoothed histogram for background segmentation. The combination of selective filling, convex hull, and Hough circle detection algorithms was utilized for the intact segmentation of red blood cells. The markers were computed from the resulted mask, and finally, marker-controlled watershed segmentation was applied to separate overlapping red blood cells. As a result, the proposed algorithm achieved higher background segmentation accuracy compared to popular background segmentation algorithms, and the inclusion of corner details improved watershed segmentation efficiency.

Automated Bacteria Colony Counting on Agar Plates Using Machine Learning

The identification of E. coli bacteria is critical for the prevention of health risks. According to EPA-approved gold standard methods, 24–48 h are required to count viable cells in water. Manual counting of viable bacteria colonies on agar plates is time-consuming and can be prone to human error. The method requires experts to identify and count colonies on agar plates using a microscope. Hence, the bacterial counting procedure must be automated in order to decrease error. The main objective of this study was to develop an automatic system for bacteria colony counting. A total of 1,301 groundwater samples were collected from eight districts in Rajasthan, India, for a field investigation. The results were validated using artificial intelligence (AI) methods on this experimental data set. We automated the process of E. coli bacteria identification using a convolutional neural network (CNN). We developed a smartphone application for the rapid detection of E. coli bacteria on agar plates using CNN. We also automated the process of bacteria colony counting using faster region-based convolutional neural network (R-CNN) to overcome manual cell counting process limitations. A graphical user interface (GUI) application was created to rapidly count bacteria colony–forming units on agar plates using faster R-CNN. The developed faster R-CNN model achieved an overall accuracy of 97% and an error (loss) of 0.10. The performance of the CNN and faster R-CNN models was validated using F-score, precision, sensitivity, and accuracy statistical measures. The comparative analysis showed that the faster R-CNN model is reliable and effective in E. coli cell counting. The study developed a system for identifying and counting viable cells of E. coli bacteria in water that can be used to forecast hotspots of water contamination.

Development of a Machine learning image segmentation-based algorithm for the determination of the adequacy of Gram-stained sputum smear images

BackgroundMachine learning (ML) prepares and trains a model through supervised or unsupervised learning methods. Sputum, a respiratory tract secretion, is a common laboratory specimen that aids in diagnosing respiratory diseases, including pulmonary tuberculosis (TB). Gram stain is an easy, cost-effective stain, which may be applied to sputum smears to screen out an unsatisfactory sample. ML model may help in screening sputum smears. MethodsThis collaborative project was carried out from June 2020-July 2021. In this study, a color-based segmentation ML algorithm using K-Means clustering was developed. A library of stained sputum smears was built. The Bartletts criteria (based on neutrophil and squamous cell count) for screening and selecting satisfactory sputum smears were used. A smartphone camera was used to take several photographs of satisfactory, as well as unsatisfactory, smears. The image segmentation algorithm was applied to medical image analysis, color-segmentation of sputum images was done. The hue saturation value (HSV) color ranges were defined on a prototype image. Then, all connected pixels were identified as a single object, and morphological operations were applied. ResultsUsage of AI-driven model on the slide-image revealed the slide adequacy as the cell count was acceptable based on Bartlett’s criteria. Both the manual cell counts (Range: 126–203 neutrophils, 14–47 squamous cells) and the model counts (Range: 117–242 neutrophils, 14–37 squamous cells) are within acceptable limits. ConclusionThe use of a model to screen a large number of sputum slides may be a boon in resource-limited settings where trained microscopists may not be easily available.

Hematologic and Biochemical Reference Intervals and Urinary Test Results for Wild-caught Adult Southern Giant Pouched Rats (Cricetomys ansorgei).

Southern giant pouched rats (Cricetomys ansorgei) are muroid rodents native to subSaharan Africa. They are increasingly used as service animals because of their keen sense of smell and are primarily known for clearing minefields in Africa. The objectives of this study were to determine hematologic and biochemical reference intervals from clinically healthy wild-caught captive adult rats, to describe the cytochemical staining reactions of peripheral blood leukocytes, and to document urinalysis findings. Blood samples were collected from the coccygeal artery of 60 isoflurane-anesthetized rats (36 males and 24 females) and analyzed with automated hematologic and biochemical analyzers; manual differential cell counts were performed on modified Wright-stained blood smears. Urine was collected by cystocentesis, and dipsticks were analyzed on a urine analyzer, with visual examination of unstained sediments. Samples from a male rat with chronic renal disease were excluded from analysis. Reference intervals were determined according to guidelines established by the American Society of Veterinary Clinical Pathology. Lymphocytes were the dominant leukocyte in peripheral blood and granular lymphocytes were identified in most animals. Male rats had significantly higher RBC, absolute reticulocyte counts, and MCV than did female rats. Minor sex-associated differences in urea nitrogen concentration and GGT activity were noted. Leukocytes showed unique cytochemical staining characteristics. Small amounts of protein and bilirubin were found in the urine of rats of both sexes and of female rats, respectively, particularly in concentrated urine. These results will provide benchmarks for determining health status and identifying disease in this species of rat.

PERFORMANCE EVALUATION OF NUCLEATED RED BLOOD CELL (NRBC) COUNT USING A FULLY AUTOMATED HAEMATOLOGY ANALYZER VERSUS MANUAL COUNTING

Objective: To evaluate the performance of Nucleated RBC (NRBC) Count using a fully automated haematology analyzer versus manual counting.
 Study Design: Cross-Sectional Study.
 Place and Duration of Study: Department of Hematology, Armed Forces Institute of Pathology, from Sep 2019-Jun 2020.
 Methodology: Routine fresh whole blood samples were run on Sysmex XN-3000 automated haematology analyzer and 384 samples with results of ≥0.1% Nucleated red blood cells were included in this study. Manual NRBC counting was carried out twice on Leishman-stained peripheral blood smears from all 384 samples. Comparison between manual and automated nucleated red blood cell counting methods was statistically analyzed through linear regression analysis & coefficient correlation. The degree of agreement between two methods was analyzed through Bland-Altman plot. Finally, concordance between the two methods was also analyzed at 5 different ranges of nucleated red blood cells.
 Results: Linear regression analysis revealed a (r2) value of 0.97. Regression equation was calculated as XN = 0.76MC ± 1.28, with 95% limits of agreement between ± 40.42% and -24.47%. A mean bias of 7.97% was demonstrated through Bland-Altman plot. Concordance analysis revealed a concordance rate of 93.74% (360/384). Nucleated red blood cell counting between two methods were more concordant when nucleated red blood cell counts were <200%.
 Conclusion: Nucleated red blood cells counting by XN-3000 automated hematology analyzer is statistically comparable to manual nucleated red blood cell counting. We suggest that automated counting can be adopted in routine hematology laboratory as a replacement of manual NRBC counting.

Machine learning-based automated fungal cell counting under a complicated background with ilastik and ImageJ.

Measuring the concentration and viability of fungal cells is an important and fundamental procedure in scientific research and industrial fermentation. In consideration of the drawbacks of manual cell counting, large quantities of fungal cells require methods that provide easy, objective and reproducible high‐throughput calculations, especially for samples in complicated backgrounds. To answer this challenge, we explored and developed an easy‐to‐use fungal cell counting pipeline that combined the machine learning‐based ilastik tool with the freeware ImageJ, as well as a conventional photomicroscope. Briefly, learning from labels provided by the user, ilastik performs segmentation and classification automatically in batch processing mode and thus discriminates fungal cells from complex backgrounds. The files processed through ilastik can be recognized by ImageJ, which can compute the numeric results with the macro ‘Fungal Cell Counter’. Taking the yeast Cryptococccus deneoformans and the filamentous fungus Pestalotiopsis microspora as examples, we observed that the customizable software algorithm reduced inter‐operator errors significantly and achieved accurate and objective results, while manual counting with a haemocytometer exhibited some errors between repeats and required more time. In summary, a convenient, rapid, reproducible and extremely low‐cost method to count yeast cells and fungal spores is described here, which can be applied to multiple kinds of eucaryotic microorganisms in genetics, cell biology and industrial fermentation.

Apoptotic Gastritis in Melanoma Patients Treated With PD-1-Based Immune Checkpoint Inhibition - Clinical and Histopathological Findings Including the Diagnostic Value of Anti-Caspase-3 Immunohistochemistry.

BackgroundGastritis induced by checkpoint inhibitors (CPI) is a rare but severe drug-related side effect. The reference standard for confirming CPI-associated gastritis (CPI-assGastritis) is histopathological assessment; however, the histopathological features of CPI-assGastritis are not yet adequately defined.Materials and MethodsGastric biopsies of melanoma patients with histopathologically suspected CPI-assGastritis were compared with gastric biopsies of patients with inflammation free gastric mucosa (IFGM), type A, B, and C gastritis with respect to apoptosis count and predominant histopathological features. Immunohistochemical anti-caspase-3 staining was performed to identify apoptosis. Quantification was performed by manually counting the number of apoptotic events per 10 high-power fields (HPF). Clinical symptoms, treatment, and follow-up data of patients with CPI-assGastritis were examined. The nonparametric Mann–Whitney U test was used for statistical testing.ResultsFive melanoma patients (three women, two men; median age: 45 years) were treated with PD-1-based CPI. The patients reported epigastric pain, weight loss, nausea, and vomiting. Histologically, the patients with CPI-assGastritis showed a partly lymphocytic, partly granulocytic inflammatory infiltrate. Manual counting of apoptotic cells in biopsy tissue slides stained against caspase 3 revealed a median of 6 apoptotic events/10 HPF (95% CI, 2.75-17.30) in the patients with CPI-assGastritis. Results for the comparison cohort (patients n = 21) were a median of 1 apoptotic event/10 HPF (95% CI, 0.5–4.5) for type-A gastritis (six patients), a median of 2 apoptotic events/10 HPF (95% CI, 0–4.5) for type-B gastritis (five patients), and no apoptosis for IFGM and type-C gastritis (five patients). Patients with CPI-assGastritis had a significantly higher apoptosis count than patients with IFGM (p<0.01), type A (p<0.05), B (p<0.05), and C gastritis (p<0.01). None of the CPI-assGastritis biopsies showed evidence of Helicobacter pylori. All CPI-assGastritis patients responded to systemic treatment with corticosteroids.ConclusionCPI-assGastritis manifests with nonspecific symptoms but histologically shows a high number of apoptotic events, which can best be visualized by anti-caspase-3 immunohistochemistry. This histopathological feature may help to histologically differentiate CPI-assGastritis from other forms of gastritis and inform decision-making regarding its optimal management.

Automatic cell counting for phase-contrast microscopic images based on a combination of Otsu and watershed segmentation method.

Cell counting plays a vital role in biomedical researches. However, manual cell counting is time-consuming, laborious, and low efficiency and has a high counting error rate problem. An automatic counting approach for Hela cells of phase-contrast microscopic images is proposed based on the combination of Otsu and watershed segmentation methods to solve the mentioned issues. Firstly, image preprocessing is performed. Secondly, the Otsu method was used to obtain an automatic global optimal threshold for segmentation to achieve batch counting of images. Thirdly, the marker watershed was performed to separate adherent cells and to avoid over-segmentation simultaneously. Finally, cells in phase-contrast microscopic images were counted by detecting the numbers of connected domains in the binary image. Taking the manual counting result as the counting standard and MIS, INC, and ACC are used as evaluation indicators. The experimental results showed that the average values of MIS, INC, and ACC of the proposed method are only 3.31%, 3.49%, and 96.69%, respectively. Additionally, each cell image was counted only takes 0.65 s on averagely. To further test the performance of the proposed method, a comparative experiment was carried out by Image J, and the result shows that the proposed method has a better counting performance with a higher average accuracy of 96.55% to Image J with 93.39%.The proposed method for cell counting is simple, feasible, fast and high accurate, and it can be used as an effective method for cell counting of the phase-contrast microscopic images.

Using Automatic Cell Counts to Calculate the I:T Ratio for Investigation of Early Onset Neonatal Sepsis

Background: Early onset neonatal sepsis is a difficult diagnosis to confirm and current practice guidelines rely heavily on the presence of risk factors and the clinician's index of suspicion1. Unfortunately, there are no laboratory tests with a strong predictive value. At our two regional institutions, newborns that are at high risk for neonatal sepsis are assessed with a CBC and the calculation of the immature to total neutrophils (I:T ratio). This calculation requires a differential count to be performed, which takes an average of 4 hours. If the results are normal, the neonate may be discharged home. In Hamilton, approximately 7,000 children are born annually, with about 1,300 of them investigated for sepsis.Our newly implemented Sysmex hematology analyzers have the ability to perform a cell count and result the absolute immature granulocyte count and absolute neutrophil count with the CBC results - these results are available at the same time as the CBC.Objective: To compare the predictive value of the calculated I:T ratio from automatic and manually counted CBC differentials in neonates with suspected early sepsis.Methods: A retrospective review of the first blood draws within the 24 hours of life on newborns being investigated for early onset sepsis over a period of 7 months at two teaching hospitals. All samples had a 100 cell count differential performed on the Cellavision as well as an automatic 32,000 cell count differential performed using the Sysmex XN series analyzers. The I:T ratio was then calculated based on the results from methods (Figure 1 and 2).Results are clinically concerning when the calculated I:T ratio is >0.2. We considered results to be discordant when only one of the two manual or automatic immature cell counts was >0.2.Results: 768 newborns were investigated for possible sepsis during this time period. 7.7% of these patients had an elevated I:T ratio calculation based on the traditional 100 cell count manual differential.The I:T calculation with the automatic results correlated with the manual in 93.2% of cases. Of the discordant 52 cases, Sysmex produced internal error flags which would trigger a manual differential to be performed and reviewed in 50 cases. A repeated manual differential reported a normal I:T ratio for one of these cases; thus only 1 of the 768 assessed samples was truly discordant.Conclusions: We found a high correlation between manual and automatic counts; in cases of discordant counts a manual count was automatically generated based on error flagging criteria. Based on our results, we are proposing an algorithm for the investigation of early onset neonatal sepsis (Figure 3).Automatic cell counts are able to correctly identify the majority of neonates with a high I:T ratio. This algorithm will reduce manual differential rates in this population by up to 93%, reducing up to 4 hours from the “usual care” of these patients.

A sample-preparation-free, automated, sample-to-answer system for cell counting in human body fluids.

While many clinical laboratory tests are now highly automated, body fluid cell counting, particularly in low-cellularity samples such as cerebral spinal fluid (CSF), is often performed manually. Here, we report a simple, cost-effective method to obtain white and red blood cell counts from human body fluids such as CSF. The method consists of a compact, automated, and low-cost fluorescence microscope system, coupled to a sample chamber containing all of the necessary reagents in dry form to stain and prepare the sample. Sample focus and scanning are handled automatically, and the acquired multimodal images are automatically analyzed to extract cell counts. Comparison with manual counting on over 200 clinical samples shows excellent agreement. As the system counts a substantially larger image region than a standard manual cell count, we find our sensitivity to extremely low cellularity samples to potentially be higher than the manual gold standard, evidenced by our system recording images of cells in samples whose cell count was registered as "0" by a trained user. Thus, our system holds promise for routine, automated, and sensitive analysis of body fluids whose cellularity extends across a wide dynamic range.

Detection of Acute Lymphoblastic Leukemia in Microscopic images using Image Processing Techniques

Leukemia is a form of blood cancer that affects the body’s ability to fight against infection. Every year, about 1 million cases of leukemia are reported with an increased mortality rate that can cause a delay in diagnosis and treatment of leukemia. Conventionally, Acute Lymphoblastic Leukemia is identified by manual counting of cells from a peripheral blood smear or by bone marrow aspiration. However, this method is time consuming and prone to human errors. To prevail over this, various automated techniques were introduced which are faster, reliable and cheaper than the manual methods. This paper focuses on how image processing techniques can be used to identify Acute Lymphoblastic Leukemia (ALL) from the peripheral blood smear images.

Machine learning models to quantify HER2 for real-time tissue image analysis in prospective clinical trials.

3061 Background: Patient eligibility for HER2-targeting treatments is commonly informed by testing tumor HER2 expression using immunohistochemistry. As HER2 expression is visually assessed by pathologists, inter- and intra-rater variability might affect treatment decisions. Here, we report the development of an automated machine learning (ML)-based algorithm to quantify HER2 cell membrane expression across a diversity of breast cancer phenotypes as a clinical tool for monitoring HER2 testing quality. Methods: A total of 689 breast cancer tissue samples were either procured (Avaden Biosciences) or were anonymized samples from the AstraZeneca biobank comprising tissues from primary and metastatic tumors, core needle biopsies and surgical resections, lobular and ductal carcinomas, across tumor grades and HER2 expression levels. Samples were stained for HER2 detection (Ventana HER2 (4B5) Assay) and digitized (Leica Biosystems) across 5 laboratories in the US. Whole-slide images (WSIs) were stratified into training (n = 407), validation (n = 110), and test sets (n = 172). Multiple convolutional neural network based ML models (PathAI, Boston, MA) were trained using 190,000 manual annotations provided by 30 board-certified pathologists to identify artifacts, invasive tumor, identify individual cancer cells and measure tumor cell membrane HER2 expression as partial or complete, and negative, weak-or-moderate, or intense. Cell-level scores were validated against a consensus of manual cell counts from 5 independent pathologists in 320 representative regions of test set WSIs. HER2 scores were generated by automatically applying rules derived from 2018 ASCO/CAP guidelines and then compared in the test set with consensus scores from 3 independent pathologists. Results: Cell counts provided by the ML model were strongly consistent with cell counts obtained by pathologist consensus in all cell-types except for faintly positive HER2 cells where ML-based quantification identified more cells on average. Automatically generated ML-ASCO/CAP HER2 scores using WSI showed substantial consistency across IHC categories with the consensus of pathologists (ICC 0.88, 95%CI 0.82-0.92) in the test set and improved further when ML models were trained to agree with pathologists by adjusting cut offs (ICC 0.91, 95%CI 0.89-0.94). The ML-based model was deployed through the PathAI cloud platform to calculate HER2 testing quality control metrics in real-time in multicentric clinical trials. Conclusions: Automated image analysis of HER2-stained breast cancer tissues using ML-based models is consistent with pathologist consensus across breast cancer tissue types. The results support evidence that ML-based algorithms can help pathologists assess HER2 testing reproducibility in clinical trials.

Automated Analysis of Cerebrospinal Fluid Cells Using Commercially Available Blood Cell Analysis Devices-A Critical Appraisal.

The analysis of cells in the cerebrospinal fluid (CSF) is a routine procedure that is usually performed manually using the Fuchs–Rosenthal chamber and cell microscopy for cell counting and differentiation. In order to reduce the requirement for manual assessment, automated analyses by devices mainly used for blood cell analysis have been also used for CSF samples. Here, we summarize the current state of investigations using these automated devices and critically review their limitations. Despite technical improvements, the lower limit for reliable leukocyte counts in the CSF is still at approximately 20 cells/µL, to be validated depending on the device. Since the critical range for clinical decisions is in the range of 5–30 cells/µL this implies that cell numbers < 30/µL require a manual confirmation. Moreover, the lower limit of reliable erythrocyte detection by automated devices is at approximately 1000/µL. However, even low erythrocyte numbers may be of clinical importance. In contrast, heavily hemorrhagic samples from neurosurgery may be counted automatically at an acceptable precision more quickly. Finally, cell differentiation by automated devices provides only a rough orientation for lymphocytes, granulocytes and monocytes. Other diagnostically important cell types such as tumor cells, siderophages, blasts and others are not reliably detected. Thus, although the automation may give a gross estimate sufficient for the emergency room situation, each CSF requires a manual microscopy for cytological evaluation for the final report. In conclusion, although automated analysis of CSF cells may provide a first orientation of the cell profile in an individual sample, an additional manual cell count and a microscopic cytology are still required and represent the gold standard.

Performance of Leukocyte Esterase Reagent Strips in the Detection of Spontaneous Bacterial Peritonitis in Cirrhotic Patients: A Systematic Review and Meta-analysis

Spontaneous bacterial peritonitis (SBP) is a bacterial infection associated with a high mortality rate in cirrhotic patients. The gold standard for the detection of SBP is a manual cell count from ascitic fluid; however, alternative screening methods are under investigation. In particular, leukocyte esterase reagent strips (LERS) has been studied as an alternative method to detect SBP with a low cost and instant turnaround time. Therefore, this study aims to evaluate the performance of LERS in the detection of SBP. A literature search was performed for studies evaluating LERS for the detection of SBP on PubMed, Embase, Scopus, Cochrane, and clinical trial registries. Summary sensitivity, specificity, log diagnostic odds ratio (LDOR), and the area under the summary receiver operating curve (AUC) were calculated according to the respective manufacturer. In total, 31 studies were evaluated. The summary sensitivity of Aution Sticks, Combur, Multistix, Periscreen reagent strips was 0.962 (95% confidence interval [CI] 0.926, 0.998), 0.892 (95% CI 0.846, 0.938), 0.806 (95% CI 0.738, 0.874), and 0.939 (95% CI 0.900, 0.979), respectively. The summary specificity of Aution Sticks, Combur, Multistix, and Periscreen reagent strips was 0.940 (95% CI 0.904, 0.976), 0.922 (95% CI 0.874, 0.970), 0.974 (95% CI 0.962, 0.985), and 0.672 (95% CI 0.381, 0.963), respectively. LERS appears to have a notable overall performance for the detection of SBP. LERS appeared to be an acceptable alternative to diagnose SBP in facilities without ability to perform cell count. However, there were significant differences in performance between each manufacturer.

Identification of Sickle Cell Anemia Using Deep Neural Networks

A molecule called hemoglobin is found in red blood cells that holds oxygen all over the body. Hemoglobin is elastic, round, and stable in a healthy human. This makes it possible to float across red blood cells. But the composition of hemoglobin is unhealthy if you have sickle cell disease. It refers to compact and bent red blood cells. The odd cells obstruct the flow of blood. It is dangerous and can result in severe discomfort, organ damage, heart strokes, and other symptoms. The human life expectancy can be shortened as well. The early identification of sickle calls will help people recognize signs that can assist antibiotics, supplements, blood transfusion, pain-relieving medications, and treatments etc. The manual assessment, diagnosis, and cell count are time consuming process and may result in misclassification and count since millions of red blood cells are in one spell. When utilizing data mining techniques such as the multilayer perceptron classifier algorithm, sickle cells can be effectively detected with high precision in the human body. The proposed approach tackles the limitations of manual research by implementing a powerful and efficient MLP (Multi-Layer Perceptron) classification algorithm that distinguishes Sickle Cell Anemia (SCA) into three classes: Normal (N), Sickle Cells(S) and Thalassemia (T) in red blood cells. This paper also presents the precision degree of the MLP classifier algorithm with other popular mining and machine learning algorithms on the dataset obtained from the Thalassemia and Sickle Cell Society (TSCS) located in Rajendra Nagar, Hyderabad, Telangana, India. Doi: 10.28991/esj-2021-01270 Full Text: PDF

Automated cell differential count in sputum is feasible and comparable to manual cell count in identifying eosinophilia

Introduction Cell differential count (CDC) of induced sputum is considered the gold standard for inflammatory phenotyping of asthma but is not implemented in routine care due to its heavy time- and staff demands. Digital Cell Morphology is a technique where digital images of cells are captured and presented preclassified as white blood cells (neutrophils, eosinophils, lymphocytes, macrophages, and unidentified) and nonwhite blood cells for review. With this study, we wanted to assess the accuracy of an automated CDC in identifying the key inflammatory cells in induced sputum. Methods Sputum from 50 patients with asthma was collected and processed using the standard processing protocol with one drop 20% albumin added to hinder cell smudging. Each slide was counted automatically using the CellaVision DM96 and manually by an experienced lab technician. Sputum was classified as eosinophilic or neutrophilic using 3% and 61% cutoffs, respectively. Results We found a good agreement using intraclass correlation for all target cells, despite significant differences in the cell count rate. The automated CDC had a sensitivity of 65%, a specificity of 93%, and a kappa-coefficient of 0.61 for identification of sputum eosinophilia. In contrast, the automated CDC had a sensitivity of 29%, a specificity of 100%, and a kappa-coefficient of 0.23 for identification of sputum neutrophilia. Conclusion Automated- and manual cell counts of sputum agree with regards to the key inflammatory cells. The automated cell count had a modest sensitivity but a high specificity for the identification of both neutrophil and eosinophil asthma.