Highly accurate differentiation of bone marrow cell morphologies using deep neural networks on a large image data set

Matek, Christian; Krappe, Sebastian; Münzenmayer, Christian; Haferlach, Torsten; Marr, Carsten

doi:10.1182/blood.2020010568

Cited by 106 publications

(94 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Most prior studies explored automation of BMA slide analysis by developing machine learningbased models for cell detection and classification [13][14][15][16][17] . However, these studies fall short of providing fully automated BMA DCCs by relying on manual selection of optimal regions for subsequent cell detection and classification, focusing on select portions of the BMA analysis pipeline, or restricting their application to non-neoplastic samples.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

An Automated Pipeline for Differential Cell Counts on Whole-Slide Bone Marrow Aspirate Smears

Lewis

Shebelut

Drumheller

et al. 2022

Preprint

View full text Add to dashboard Cite

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.

show abstract

Section: Discussionmentioning

confidence: 99%

“…Fu et al compared automated classification results to manual DCCs and obtained strong correlation for three cell types 15 . Matek et al and Yu et al developed large expert-annotated training datasets to train highly accurate CNN-based cell classification models 16,17 .…”

Section: Introductionmentioning

confidence: 99%

An Automated Pipeline for Differential Cell Counts on Whole-Slide Bone Marrow Aspirate Smears

Lewis

Shebelut

Drumheller

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…This may be helpful to identify patients in need of further and usually more invasive and expensive testing, such as bone marrow aspirates or genome sequencing. Recent applications of computational cytomorphology on bone marrow smears have demonstrated its ability to automatically identify different leukocytes 30,62 and assist diagnostic predictions 27–29 in specialized haemato-oncology. By demonstrating that this can now be extended to blood smears/slides, our work reveals the potential for the large-scale incorporation of automated cytomorphology into routine diagnostic workflows.…”

Section: Discussionmentioning

confidence: 99%

“…This can create challenges in identifying relevant cytomorphology-disease associations. Computational methods, which have shown promise in the characterization and prediction of MDS and AML using bone marrow slides 27–29 and identification of abnormal leukocytes 30 , can help address these problems.…”

Section: Introductionmentioning

confidence: 99%

Computational analysis of peripheral blood smears detects disease-associated cytomorphologies

Almeida

Gudgin

Besser

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Many hematological diseases are characterized by altered abundance and morphology of blood cells and their progenitors. Myelodysplastic syndromes (MDS), for example, are a type of blood cancer manifesting via a range of cytopenias and dysplastic changes of blood and bone marrow cells. While experts analyze cytomorphology to diagnose MDS, similar alterations can be observed in other conditions such as haematinic deficiency anemias, and definitive diagnosis requires complementary information such as blood counts, karyotype and molecular testing. However, recent works demonstrated that computational analysis of bone marrow slides predicts not only MDS or AML but also the presence of specific mutations. Here, we present and make available Haemorasis, a computational method that detects and characterizes white and red blood cells (WBC and RBC, respectively) in peripheral blood slides, and apply it to over 300 individuals with different conditions (SF3B1-mutant and SF3B1-wildtype MDS, megaloblastic anemia and iron deficiency anemia), where Haemorasis detects over half a million WBC and millions of RBC. We then show how these large sets of cell images can be used in diagnosis and prognosis, whilst identifying novel associations between computational morphotypes and disease. We find that hypolobulated neutrophils and large RBC are characteristic of SF3B1-mutant MDS, and, while prevalent in both iron deficiency and megaloblastic anemia, hyperlobulated neutrophils are larger in the latter. Finally, we externally validate these methods, showing they generalize to other centers and scanners.

show abstract

“…The dataset, acquired by Matek et al, contains 171,375 images from a cohort of 945 patients diagnosed with various hematological diseases at MLL Munich Leukemia Laboratory [51]. The minimum patient age was 18.1 years, and the maximum was 92.2 years.…”

Section: Methodsmentioning

confidence: 99%

HematoNet: Expert Level Classification of Bone Marrow Cytology Morphology in Hematological Malignancy with Deep Learning

Tripathi

Augustin

Sukumaran

et al. 2022

Preprint

View full text Add to dashboard Cite

There have been few efforts made to automate the cytomorphological categorization of bone marrow cells. For bone marrow cell categorization, deep-learning algorithms have been limited to a small number of samples or disease classifications. In this paper, we proposed a pipeline to classify the bone marrow cells despite these limitations. Data augmentation was used throughout the data to resolve any class imbalances. Then, random transformations such as rotating between 0° to 90°, zooming in/out, flipping horizontally and/or vertically, and translating were performed. The model used in the pipeline was a CoAtNet and that was compared with two baseline models, EfficientNetV2 and ResNext50. We then analyzed the CoAtNet model using SmoothGrad and Grad-CAM, two recently developed algorithms that have been shown to meet the fundamental requirements for explainability methods. After evaluating all three models’ performance for each of the distinct morphological classes, the proposed CoAtNet model was able to outperform the EfficientNetV2 and ResNext50 models due to its attention network property that increased the learning curve for the algorithm which was represented using a precision-recall curve.

show abstract

Highly accurate differentiation of bone marrow cell morphologies using deep neural networks on a large image data set

Cited by 106 publications

References 42 publications

An Automated Pipeline for Differential Cell Counts on Whole-Slide Bone Marrow Aspirate Smears

An Automated Pipeline for Differential Cell Counts on Whole-Slide Bone Marrow Aspirate Smears

Computational analysis of peripheral blood smears detects disease-associated cytomorphologies

HematoNet: Expert Level Classification of Bone Marrow Cytology Morphology in Hematological Malignancy with Deep Learning

Contact Info

Product

Resources

About