Search

Pathologic diagnosis of bone marrow disorders relies in part on microscopic analysis of bone marrow aspirate (BMA) smears and manual counting of marrow nucleated cells to obtain a differential cell count (DCC). This manual process has significant limitations, including analysis of only a small subset of optimal slide areas and nucleated cells, and inter-observer variability due to differences in cell selection and classification. To address these shortcomings, we developed an automated machine learning-based pipeline for obtaining 11-component DCCs on whole-slide BMAs. This pipeline utilizes a sequential process of identifying optimal BMA regions with high proportions of marrow nucleated cells, detecting individual cells within these optimal areas, and classifying these cells into one of 11 DCC components. Convolutional neural network models were trained on 396,048 BMA region, 28,914 cell boundary, and 1,510,976 cell class images from manual annotations. The resulting automated pipeline produces 11-component DCCs that demonstrate high statistical and diagnostic concordance with manual DCCs among a heterogeneous group of testing BMA slides with varying pathologies and cellularities. Additionally, we show that automated analysis can reduce intra-slide variance in DCCs by analyzing the whole slide and marrow nucleated cells within optimal regions. Finally, pipeline outputs of region classification, cell detection, and cell classification can be visualized using whole-slide image analysis software. This study demonstrates the feasibility of a fully-automated pipeline for generating DCCs on scanned whole-slide BMA images, with the potential for improving the current standard of practice for utilizing BMA smears in the laboratory analysis of hematologic disorders.

Current pathologic diagnosis of benign and neoplastic bone marrow disorders relies in part on the microscopic analysis of bone marrow aspirate (BMA) smears and manual counting of nucleated cell populations to obtain a cell differential. This manual process has significant limitations, including the limited sample of cells analyzed by a conventional 500-cell differential compared to the thousands of nucleated cells present, as well as the inter-observer variability seen between differentials on single samples due to differences in cell selection and classification. To address these shortcomings, we developed an automated computational platform for obtaining cell differentials from scanned whole-slide BMAs at 40x magnification. This pipeline utilizes a sequential process of identifying BMA regions with high proportions of marrow nucleated cells that are ideal for cell counting, detecting individual cells within these optimal regions, and classifying cells into one of 11 types within the differential. Training of convolutional neural network models for region and cell classification, as well as a region-based convolutional neural network for cell detection, involved the generation of an annotated training data set containing 10,948 BMA regions, 28,914 cell boundaries, and 23,609 cell classifications from 73 BMA slides. Among 44 testing BMA slides, an average of 19,209 viable cells per slide were identified and used in automated cell differentials, with a range of 237 to 126,483 cells. In comparing these automated cell differential percentages with corresponding manual differentials, cell type-specific correlation coefficients ranged from 0.913 for blast cells to 0.365 for myelocytes, with an average coefficient of 0.654 among all 11 cell types. A statistically significant concordance was observed among slides with blast percentages less or greater than 20% (p=1.0x10-5) and with plasma cell percentages less or greater than 10% (p=5.9x10-6) between automated and manual differentials, suggesting potential diagnostic utility of this automated pipeline for malignancies such as acute myeloid leukemia and multiple myeloma. Additionally, by simulating the manual counting of 500 cells within localized areas of a BMA slide and iterating over all optimal slide locations, we quantified the inter-observer variability associated with limited sample size in traditional BMA cell counting. Localized differentials exemplify an average variance ranging from 24.1% for erythroid precursors to 1.8% for basophils. Variance in localized differentials of up to 44.8% for blast cells and 36.9% for plasma cells was observed, demonstrating that sample classification based on diagnostic thresholds of cell populations is variable even between different areas within a single slide. Finally, pipeline outputs of region classification, cell detection, cell classification, and localized cell differentials can be visualized using whole-slide image analysis software. By improving cell sampling and reducing inter-observer variability, this automated pipeline has potential to improve the current standard of practice for utilizing BMA smears in the diagnosis of hematologic disorders.

Lymphoid follicles/aggregates in gastric biopsies have been traditionally linked to Helicobacter pylori gastritis, and less commonly to other inflammatory and neoplastic conditions. The frequency of such aggregates in normal stomachs has yet to be adequately evaluated. This is especially relevant when it comes to diagnosing non-specific chronic gastritis in biopsy specimens with chronic inflammation but no evidence of H pylori infection. Sleeve gastrectomies represent an opportunity to study adequately preserved gastric mucosa in patients who are otherwise asymptomatic and lack a history of gastric disease.To study sleeve gastrectomy specimens to quantify the amount of lymphoid follicles/aggregates and lymphocytic infiltration in normal stomachs.Sixty-eight bariatric sleeve gastrectomies and 13 control specimens from Whipple resections were examined for multiple histologic features including type, quantity, and distribution of chronic inflammation and lymphoid follicles/aggregates. Presence of H pylori was documented by both Hematoxylin and eosin-stained (H&E) and immunohistochemistry (IHC). Clinical information including age, sex, medication intake, prior endoscopy, and/or H pylori infection was recorded. The patient population was divided in 2 groups, H pylori negative versus H pylori positive, and statistical analysis was performed by a biostatistician.Two hundred sixty three fundic sections from 68 bariatric patients were examined. Fifty three patients were found to be H pylori-negative, compared with 15 who were positive for H pylori. Among the H pylori-negative group, the average number of lymphoid aggregates was 3.33, compared with an average of 6.26 in the H pylori positive group (the difference was statistically significant with a P-value of .008). The average number of plasma cells per high power field was 2.15 in the H pylori negative group, compared and average of 5.07 in the H pylori positive group (the difference was also statistically significant with a P-value

CD30 is a member of the tumor necrosis factor family of cell surface receptors normally expressed in lymphocytes, as well as some lymphomas, but has been described in other malignancies. Anaplastic lymphoma kinase (ALK) is a tyrosine kinase receptor that belongs to the insulin receptor superfamily, and is normally expressed in neural cells, but has been detected in several malignancies. There is conflicting data in the literature that describes the expression of these receptors in breast cancer, and the aim of this study is to test the expression of CD30 and ALK in a cohort of Middle Eastern patients with breast carcinoma. Cases of invasive breast cancer from the archives of AUBMC were reviewed over a period of 9 years, and the blocks that were used for immunohistochemical staining for ER, PR, Her-2/neu were selected. Immunohistochemical staining for CD30 (JCM182) and ALK (5A4 and D5F3) was performed. Two hundred eighty-four cases were identified (2 cases were male), with a mean age of 55 ± 12. CD30 and ALK expression was not seen in any of the cases. Our cohort showed complete negativity to both CD30 and ALK, adding to the conflicting data available in the literature, and more studies are needed to reliably identify a trend of expression of CD30 and ALK in breast carcinoma, especially in the Middle East.

Introduction: Nowadays, medical treatments efficiency is challenged by multi drug resistant organisms (MDROs). Lebanese nursing homes’ residents revealed high fecal carriage rates of MDR Enterobacteriacea. Previous studies claim that bacteria with resistant genes experience fitness cost. This study assesses the competitive growth of MDR Escherichia coli and Klebsiella pneumoniae. Methodology: Fecal swabs were collected, during six consecutive months, from ten elderly residing in a Lebanese nursing home. All isolates were subject to API 20E (bioMerieux, Marcy L’ Etoile, France) and antimicrobial susceptibility (Kirby–Bauer method) testing. Phenotypically, ESBL (extended spectrum β-lactamase), MBL (metallo β-lactamase), AmpC and KPC (Klebsiella pneumoniae carbapenemase) were detected using EDTA, PBA, cloxacillin, and DDSTs (Biorad, Hercules, USA). Selected ESBL producing E. coli and K. pneumoniae underwent multiplex PCR analysis. Intra and inter-species in-vitro competition assays were conducted in multiple combinations. Results: Among 117 collected isolates, E. coli was predominant (71.8%); 7.7% were ESBL and 5.1% AmpC producers. With E. coli intra-species assays, sensitive isolates out-competed all others, followed by ESBL, AmpC, and OXA-48 (oxacillin) producers. Inter-species assays, demonstrated a decreased fitness of ESBL producing K. pneumoniae in presence of sensitive E. coli. While out-competing ESBL producing E. coli required 2 sensitive K. pneumoniae isolates. Conclusion: This study highlights resistant E. coli and K. pneumoniae frequency decrease in presence of sensitive isolates, endorsing the fitness cost hypothesis. Hence, competing supplementary species reproducing gut flora, would ensure further steps in the fight against MDROs.