Morphogo: An Automatic Bone Marrow Cell Classification System on Digital Images Analyzed by Artificial Intelligence

2020 ◽  
Vol 64 (6) ◽  
pp. 588-596
Author(s):  
Xinyan Fu ◽  
May Fu ◽  
Qiang Li ◽  
Xiangui Peng ◽  
Ju Lu ◽  
...  
Keyword(s):  
Artificial Intelligence ◽  
Bone Marrow ◽  
Sensitivity And Specificity ◽  
Bone Marrow Aspirate ◽  
Marrow Cell ◽  
Clinical Applications ◽  
Differential Count ◽  
Cell Classification ◽  
Cell Percentage ◽  
Differential Cell

<b><i>Introduction:</i></b> The nucleated-cell differential count on the bone marrow aspirate smears is required for the clinical diagnosis of hematological malignancy. Manual bone marrow differential count is time consuming and lacks consistency. In this study, a novel artificial intelligence (AI)-based system was developed to perform cell automatic classification of bone marrow cells and determine its potential clinical applications. <b><i>Materials and Methods:</i></b> Bone marrow aspirate smears were collected from the Xinqiao Hospital of Army Medical University. First, an automated analysis system (<i>Morphogo</i>) scanned and generated whole digital images of bone marrow smears. Then, the nucleated marrow cells in the selected areas of the smears at a magnification of ×1,000 were analyzed by the software utilizing an AI-based platform. The cell classification results were further reviewed and confirmed independently by 2 experienced pathologists. The automatic cell classification performance of the system was evaluated using 3 categories: accuracy, sensitivity, and specificity. Correlation coefficients and linear regression equations between automatic cell classification by the AI-based system and concurrent manual differential count were calculated. <b><i>Results:</i></b> In 230 cases, the classification accuracy was above 85.7% for hematopoietic lineage cells. Averages of sensitivity and specificity of the system were found to be 69.4 and 97.2%, respectively. The differential cell percentage of the automated count based on 200–500 cell counts was correlated with differential cell percentage provided by the pathologists for granulocytes, erythrocytes, and lymphocytes (<i>r</i> ≥ 0.762, <i>p</i> &#x3c; 0.001). <b><i>Discussion/Conclusion:</i></b> This pilot study confirmed that the <i>Morphogo</i> system is a reliable tool for automatic bone marrow cell differential count analysis and has potential for clinical applications. Current ongoing large-scale multicenter validation studies will provide more information to further confirm the clinical utility of the system.

scholarly journals Developing and Preliminary Validating an Automatic Cell Classification System for Bone Marrow Smears: a Pilot Study

2020 ◽  
Vol 44 (10) ◽  
Author(s):  
Hong Jin ◽  
Xinyan Fu ◽  
Xinyi Cao ◽  
Mingxia Sun ◽  
Xiaofen Wang ◽  
...  
Keyword(s):  
Bone Marrow ◽  
Cell Count ◽  
Marrow Cell ◽  
Differential Count ◽  
Cell Classification ◽  
Initial Development ◽  
Differential Cell Count ◽  
Analysis Methods ◽  
Differential Cell ◽  
Cell Series

Abstract Bone marrow smear examination is an indispensable diagnostic tool in the evaluation of hematological diseases, but the process of manual differential count is labor extensive. In this study, we developed an automatic system with integrated scanning hardware and machine learning-based software to perform differential cell count on bone marrow smears to assist diagnosis. The initial development of the artificial neural network was based on 3000 marrow smear samples retrospectively archived from Sir Run Run Shaw Hospital affiliated to Zhejiang University School of Medicine between June 2016 and December 2018. The preliminary field validating test of the system was based on 124 marrow smears newly collected from the Second Affiliated Hospital of Harbin Medical University between April 2019 and November 2019. The study was performed in parallel of machine automatic recognition with conventional manual differential count by pathologists using the microscope. We selected representative 600,000 marrow cell images as training set of the algorithm, followed by random captured 30,867 cell images for validation. In validation, the overall accuracy of automatic cell classification was 90.1% (95% CI, 89.8–90.5%). In a preliminary field validating test, the reliability coefficient (ICC) of cell series proportion between the two analysis methods were high (ICC ≥ 0.883, P < 0.0001) and the results by the two analysis methods were consistent for granulocytes and erythrocytes. The system was effective in cell classification and differential cell count on marrow smears. It provides a useful digital tool in the screening and evaluation of various hematological disorders.

scholarly journals Quantitative analysis of bone marrow in pancytopenic dogs

2017 ◽  
Vol 38 (6) ◽  
pp. 3639 ◽  
Author(s):  
Angela Ferronato Girardi ◽  
Amanda Noéli da Silva Campos ◽  
Caroline Argenta Pescador ◽  
Arleana Do Bom Parto Ferreira de Almeida ◽  
Adriane Jorge Mendonça ◽  
...  
Keyword(s):  
Bone Marrow ◽  
Bone Marrow Aspirate ◽  
Differential Count ◽  
Blood Cell Count ◽  
Chi Square ◽  
Study Region ◽  
Bone Marrow Change ◽  
Erythroid Hyperplasia ◽  
Quantitative Characteristics ◽  
Clinical Changes

The pancytopenia can be associated with intra and extra medullary disorders. When the etiology is not obvious, the examination of bone marrow is necessary. The study aims to report and discuss quantitative amendments in bone marrow and their causes in dogs with pancytopenia. Bone marrow aspirate was obtained from 65 dogs with pancytopenia over a period of 13 consecutive months for preparation of smears stained by Giemsa and observed in light microscopy. Five hundred cells differential count was held; and assessed myeloid:erythroid ratio, cellularity, megakaryocytes and direct parasitological examination. The data were evaluated by the Chi-square statistical test. Blood tests of 3120 canines with several clinical changes were analyzed to identify pancytopenia, which accounted for 167 (5.4%) dogs. Interpretation of quantitative characteristics was carried out from the bone marrow smear in 65 pancytopenic dogs and the etiology was established in 40 (61.5%) of these, which included infection by E. canis and L. chagasi, idiopathic aplastic anemia, chronic renal failure and co-infections. In 17 (26.2%) animals were not observed medullary changes. The most bone marrow change was observed myeloid and erythroid hypoplasia in 17 (26.2%) dogs, followed by myeloid and erythroid hyperplasia (24.6%). The myelogram association to the blood cell count allowed the identification of medullary findings and its participation on the occurrence of pancytopenia cases. The number of cases resulting from infectious diseases was 38 (58.5%) in canine assessed: significant number due to its endemic characteristic of the study region.

scholarly journals A systematic review of the concept and clinical applications of Bone Marrow Aspirate Concentrate in Orthopaedics

SICOT-J ◽  
2017 ◽  
Vol 3 ◽  
pp. 17 ◽  
Author(s):  
Mohamed A. Imam ◽  
Samer S.S. Mahmoud ◽  
James Holton ◽  
Dalia Abouelmaati ◽  
Yasser Elsherbini ◽  
...  
Keyword(s):  
Systematic Review ◽  
Bone Marrow ◽  
Bone Marrow Aspirate ◽  
Clinical Applications ◽  
Bone Marrow Aspirate Concentrate

scholarly journals Possible automatic cell classification of bone marrow aspirate using the CELL-DYN 4000®automatic blood cell analyzer

2002 ◽  
Vol 16 (2) ◽  
pp. 86-90 ◽  
Author(s):  
Ryousuke Yamamura ◽  
Takahisa Yamane ◽  
Masayuki Hino ◽  
Kensuke Ohta ◽  
Hisako Shibata ◽  
...  
Keyword(s):  
Bone Marrow ◽  
Blood Cell ◽  
Bone Marrow Aspirate ◽  
Cell Classification ◽  
Cell Analyzer

The Colchicine Method in the Study of Bone Marrow Cell Proliferation

Blood ◽  
1961 ◽  
Vol 18 (3) ◽  
pp. 328-335 ◽  
Author(s):  
GIUSEPPE CARDINALI ◽  
GIULIANA CARDINALI ◽  
MARIA F. AGRIFOGLIO
Keyword(s):  
Bone Marrow ◽  
Bone Marrow Cells ◽  
Marrow Cell ◽  
Differential Count ◽  
Limiting Factor ◽  
Stathmokinetic Effect ◽  
Akr Mice ◽  
Metaphase Accumulation ◽  
Mature Bone ◽  
Marrow Cells

Abstract The rate of metaphase accumulation in the bone marrow cells of AKR mice treated with colchicine was investigated. The influence of this alkaloid on the differential count of the bone marrow cells in these animals was also studied. It was demonstrated that the stathmokinetic effect of colchicine on the bone marrow cells started almost immediately after the administration of the drug. The number of arrested metaphases increased linearly from one-half hour to six hours after the injection of colchicine, and then fell rapidly. In rats injected with colchicine, the changes in the bone marrow concentration of this compound were followed for eight hours. The colchicine concentrations increased from the first to the fourth hour, and then fell rapidly, reaching the zero level at the eighth hour. From the results obtained, it appeared that four hours post-injection was the most convenient time for the study of the bone marrow proliferative activity by the colchicine method when 1.2 mg./Kg. of colchicine per body weight was used. The fact that four hours after the injection, a mild decrease in the percentage of mature bone marrow granulocytes was found, may represent a limiting factor, which is, however, of moderate importance in the application of this technic.

scholarly journals Presumptive pure erythroid leukemia in a dog

2012 ◽  
Vol 24 (5) ◽  
pp. 1004-1007 ◽  
Author(s):  
Mathios E. Mylonakis ◽  
Maria Kritsepi-Konstantinou ◽  
William Vernau ◽  
Victor E. Valli ◽  
Dimitra Pardali ◽  
...  
Keyword(s):  
Bone Marrow ◽  
Bone Marrow Aspirate ◽  
Intact Male ◽  
Cell Percentage ◽  
History Of ◽  
Blast Cells ◽  
Monomorphic Population ◽  
Cluster Of Differentiation ◽  
History Of Depression ◽  
Individual Particles

A 6.5-year-old, intact male Cocker Spaniel dog was referred with a history of depression and anorexia of 1-week duration. Mucosal pallor was prominent on physical examination. Complete blood cell count revealed pancytopenia and occasional blast cells. Bone marrow aspirate cytology indicated that individual particles were composed of approximately 60% hematopoietic cells and a monomorphic population of blast cells with perfectly round nuclei, consistent paranuclear clearing, and deeply basophilic cytoplasm devoid of granules dominating the marrow fields. The granulocytic lineage was severely decreased with a granulocytic-to-erythroid ratio of 0.15 and a blast cell percentage of at least 70% of all nucleated cells; the myeloblasts and monoblasts composed <5% of nonerythroid cells. Bone marrow cytology slides were submitted for immunocytochemical immunophenotyping using antibodies to myeloperoxidase, cluster of differentiation (CD)3, CD79a, CD11b, CD45, and CD34. The neoplastic cells did not express any of the antigens assessed. The combination of light microscopic cytomorphology and the immunophenotype were strongly suggestive of pure erythroid leukemia.

scholarly journals A systematic review of the concept and clinical applications of bone marrow aspirate concentrate in tendon pathology

SICOT-J ◽  
2017 ◽  
Vol 3 ◽  
pp. 58 ◽  
Author(s):  
Mohamed A. Imam ◽  
James Holton ◽  
Saman Horriat ◽  
Ahmed S. Negida ◽  
Florian Grubhofer ◽  
...  
Keyword(s):  
Systematic Review ◽  
Bone Marrow ◽  
Bone Marrow Aspirate ◽  
Clinical Applications ◽  
Bone Marrow Aspirate Concentrate ◽  
Tendon Pathology

A systematic review of the clinical applications and complications of bone marrow aspirate concentrate in management of bone defects and nonunions

2017 ◽  
Vol 41 (11) ◽  
pp. 2213-2220 ◽  
Author(s):  
Mohamed A. Imam ◽  
James Holton ◽  
Lukas Ernstbrunner ◽  
Wojciech Pepke ◽  
Florian Grubhofer ◽  
...  
Keyword(s):  
Systematic Review ◽  
Bone Marrow ◽  
Bone Marrow Aspirate ◽  
Bone Defects ◽  
Clinical Applications ◽  
Bone Marrow Aspirate Concentrate

Utility of CD81 Flow Cytometry (FC) Assessment In Diagnostic Evaluation of Plasma Cell (PC) Disorders

Blood ◽  
2010 ◽  
Vol 116 (21) ◽  
pp. 3011-3011
Author(s):  
Haris Kartsios ◽  
Vassilios Papadopoulos ◽  
Smaragda Effraimidou ◽  
Kyriaki Kokoviadou ◽  
Anastasia Spyrou ◽  
...  
Keyword(s):  
Multiple Myeloma ◽  
Bone Marrow ◽  
Flow Cytometry ◽  
Sensitivity And Specificity ◽  
Plasma Cell ◽  
Bone Marrow Aspirate ◽  
False Negative ◽  
Reactive Plasmacytosis ◽  
Cd56 Expression ◽  
Cd19 Expression

Abstract Abstract 3011 Background: CD81 is a tetraspanin cell surface protein that regulates CD19 expression in B lymphocytes and enables hepatitis C virus infection of human cells. Immunohistochemistry and FC have showed that CD81 expression is downregulated in multiple myeloma and PC lines (Luo, 2010). It appears also that normal PCs are CD81+ at a high percentage (Rawstron, 2008). However, information about CD81 utility in phenotypic diagnosis of PC disorders is scanty. We assessed the frequency and the diagnostic value of the FC detection of CD81 surface expression in multiple myeloma (MM), monoclonal gammopathy of undetermined significance (MGUS) and non-hematological malignancy subjects. Patients-Methods: A total of 84 bone marrow aspirates were analyzed in this study. These included 52 diagnosis/disease progression MM samples, 8 samples from MGUS patients and 24 samples from reactive plasmacytosis and ITP patients. A 4-color FC technique was used for PC analysis. Based on the European Myeloma Network guidelines (Rawstron, 2008), we evaluated expression of CD19, CD27, CD56 and CD81 on PCs identified by the CD38/CD45/CD138 combination. At least 100,000 events/tube were acquired, leading to a sensitivity limit of 0.1%. FC positivity threshold for studied antigens was set at 20%. Any deviation from the normal PC phenotype (CD19+CD27+CD56-) was considered as abnormal. Diagnostic tests also included FBC, biochemistry, beta-2 microglobulin, IgA, IgG, IgM, kappa and lambda light chain serum levels, serum protein electrophoresis and immunofixation, bone marrow aspirate and skeletal survey where required. Results: Median PC percentage was significantly higher in the bone marrow aspirate when compared to FC as expected [15.5(3-96)% vs. 2(0.11-53)%, p<0.001]. CD81 positivity was identified in 19% (10/52) of MM samples and 50% of MGUS (4/8) and normal PC samples (12/24). CD81 plasma cell expression was significantly different between normal and MM samples (median: 16.8(1.9-52.4)% vs. 2.5(0-90.4)%, p<0.001) and between MGUS and MM patients (median: 14.4(1.5-53)% vs. 2.5(0-90.4)%, p=0.001). Mean fluorescence intensity of CD81 was similar in the 3 groups. A positive correlation was found between CD81 and CD19 expression (rho=+0.596, p<0.0001) and between CD81 and CD27 (rho=+0.358, p=0.001) while CD56 expression was negatively correlated with CD81 (rho=-0.291, p=0.007). Flow cytometry provided correct diagnosis in 87% (73/84) of cases; 8 samples were identified as false positive and 3 as false negative. Subsequently, a Receiver Operating Characteristic (ROC) curve analysis was performed in order to assess CD19, CD27, CD56 and CD81 antigens' sensitivity and specificity in discriminating normal from malignant PCs (Figure); although CD81 shared similar sensitivity and specificity with CD27, CD19 was shown to have the highest sensitivity and specificity in detecting correctly PC identity. Conclusions: This study provides laboratory evidence that CD81 expression in PCs is strongly correlated with CD19, CD27 and CD56 expression. CD81 surface detection seems to characterize normal PCs in reactive plasmacytosis, ITP and MGUS. However, it has an inferior sensitivity and specificity, in discriminating normal from malignant PCs, when compared to CD19. Disclosures: No relevant conflicts of interest to declare.

Sign in / Sign up

Export Citation Format

Share Document