Dataset Information

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

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?

SUBMITTER: Jin H

PROVIDER: S-EPMC7476995 | biostudies-literature | 2020 Sep

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

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

Jin Hong H Fu Xinyan X Cao Xinyi X Sun Mingxia M Wang Xiaofen X Zhong Yuhong Y Yang Suwen S Qi Chao C Peng Bo B He Xin X He Fei F Jiang Yongfang Y Gao Haiyan H Li Shun S Huang Zhen Z Li Qiang Q Fang Fengqi F Zhang Jun J

Journal of medical systems 20200907 10

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 Si ...[more]

PMID: 32894360

Similar Datasets

Project description:BackgroundAcute promyelocytic leukemia (APL) is considered a hematologic emergency due to high risk of bleeding and fatal hemorrhages being a major cause of death. Despite lower death rates reported from clinical trials, patient registry data suggest an early death rate of 20%, especially for elderly and frail patients. Therefore, reliable diagnosis is required as treatment with differentiation-inducing agents leads to cure in the majority of patients. However, diagnosis commonly relies on cytomorphology and genetic confirmation of the pathognomonic t(15;17). Yet, the latter is more time consuming and in some regions unavailable.MethodsIn recent years, deep learning (DL) has been evaluated for medical image recognition showing outstanding capabilities in analyzing large amounts of image data and provides reliable classification results. We developed a multi-stage DL platform that automatically reads images of bone marrow smears, accurately segments cells, and subsequently predicts APL using image data only. We retrospectively identified 51 APL patients from previous multicenter trials and compared them to 1048 non-APL acute myeloid leukemia (AML) patients and 236 healthy bone marrow donor samples, respectively.ResultsOur DL platform segments bone marrow cells with a mean average precision and a mean average recall of both 0.97. Further, it achieves high accuracy in detecting APL by distinguishing between APL and non-APL AML as well as APL and healthy donors with an area under the receiver operating characteristic of 0.8575 and 0.9585, respectively, using visual image data only.ConclusionsOur study underlines not only the feasibility of DL to detect distinct morphologies that accompany a cytogenetic aberration like t(15;17) in APL, but also shows the capability of DL to abstract information from a small medical data set, i. e. 51 APL patients, and infer correct predictions. This demonstrates the suitability of DL to assist in the diagnosis of rare cancer entities. As our DL platform predicts APL from bone marrow smear images alone, this may be used to diagnose APL in regions were molecular or cytogenetic subtyping is not routinely available and raise attention to suspected cases of APL for expert evaluation.

Project description:After fractionation of rabbit bone marrow into dividing (early) and non-dividing (late) erythroid cells, the adenylate cyclase activity of membrane ghosts was assayed in the presence of guanine nucleotides ((GTP and its analogue p[NH]ppG (guanosine 5'-[beta, gamma-imido]triphosphate))), the beta-adrenergic agonist L-isoprenaline (L-isoproterenol) and the antagonist L-propranolol. Both GTP and p[NH]ppG increased the adenylate cyclase activity of early and late erythroblasts, whereas the stimulating effect of the beta-adrenergic drug L-isoprenaline was limited to the immature dividing bone-marrow cells. The effect of L-isoprenaline was completely inhibited by the antagonist L-propranolol, confirming that the response was due to stimulation of beta-adrenergic receptors on the plasma membrane. The lack of response of non-dividing erythroblasts to beta-adrenergic stimuli is not due to loss of beta-receptors, since both dividing and non-dividing cells bind the selective ligand [125I]iodohydroxybenzylpindolol with almost equal affinities, the apparent dissociation constants, Kd, being 0.91 X 10(-8)M and 1.0 X 10(-8) M respectively. The number of beta-adrenergic receptors per cell was 2-fold higher in the dividing cells. No significant change in binding affinity for GTP and p[NH]ppG during erythroblast development was observed: the dissociation constants of both guanine nucleotides were almost identical with early and late erythroblast membrane preparations [2-3 (X 10(-7) M]. With dividing cells, however, in the presence of L-isoprenaline the dissociation constants of GTP and p[NH]ppG were lower (6 X 10(-8) M). The dose-response curves for isoprenaline competition in binding of [125I]iodohydroxybenzylpindolol by dividing cells showed that the EC50 (effective concentration for half maximum activity) value for isoprenaline was higher in the presence of p[NH]ppG. With non-dividing cells the EC50 value for isoprenaline was equal in the presence and in the absence of p[NH]ppG and similar to that observed with dividing-cell membranes in the presence of the nucleotide. Thus differentiation of rabbit bone-marrow erythroid cells seems to be accompanied by uncoupling of the beta-adrenergic receptors from the adenylate cyclase catalytic protein as well as by a decrease in the number of receptors per cell, but not by changes in the catecholamine and guanine-nucleotide-binding affinities.

Dataset Information

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

Publications

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

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets