Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Long-Term Cognitive Impairment After CAR-T Therapy Versus Autologous Stem Cell Transplantation: A Propensity Score-Matched Cohort Study.

Diagnostics (Basel, Switzerland)·2026
Same author

Effects of Resveratrol on MCP-1/CCL2-Related Readouts in Preclinical Animal Models: A Systematic Review and Meta-Analysis.

Biomedicines·2026
Same author

COVID-19 mRNA vaccination and immune-related adverse events in patients with cancer receiving immune checkpoint inhibitors: a target trial emulation study.

Journal for immunotherapy of cancer·2026
Same author

Target Trial Emulation of Second-Generation Versus First-Generation Tyrosine Kinase Inhibitors in Adult Philadelphia Chromosome-Positive Acute Lymphoblastic Leukemia.

JCO oncology practice·2026
Same author

Survival, Toxicity, and Economic Outcomes of Osimertinib Versus Second-Generation Epidermal Growth Factor Receptor-Tyrosine Kinase Inhibitors in Metastatic Epidermal Growth Factor Receptor-Mutant Non-Small Cell Lung Cancer.

JCO oncology practice·2026
Same author

Sequential Treatment with Regorafenib and Trifluridine/Tipiracil in Refractory Metastatic Colorectal Cancer.

Life (Basel, Switzerland)·2026

Related Experiment Video

Updated: Dec 24, 2025

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone
09:31

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone

Published on: April 8, 2015

11.9K

A Hematologist-Level Deep Learning Algorithm (BMSNet) for Assessing the Morphologies of Single Nuclear Balls in Bone

Yi-Ying Wu1, Tzu-Chuan Huang1, Ren-Hua Ye1

  • 1Division of Hematology/Oncology, Department of Medicine, Tri-Service General Hospital, National Defense Medical Center, Taipei, Taiwan.

JMIR Medical Informatics
|April 9, 2020
PubMed
Summary

A new deep learning model, BMSNet, aids hematologists in interpreting bone marrow smears, accelerating diagnosis. While effective, expert human review remains crucial for detailed morphological analysis.

Keywords:
artificial intelligencebone marrow examinationdeep learningleukemiamyelodysplastic syndrome

More Related Videos

Analyzing Mitochondrial Morphology Through Simulation Supervised Learning
12:06

Analyzing Mitochondrial Morphology Through Simulation Supervised Learning

Published on: March 3, 2023

4.6K
Database-guided Flow-cytometry for Evaluation of Bone Marrow Myeloid Cell Maturation
12:05

Database-guided Flow-cytometry for Evaluation of Bone Marrow Myeloid Cell Maturation

Published on: November 3, 2018

12.2K

Related Experiment Videos

Last Updated: Dec 24, 2025

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone
09:31

Automated Quantification of Hematopoietic Cell – Stromal Cell Interactions in Histological Images of Undecalcified Bone

Published on: April 8, 2015

11.9K
Analyzing Mitochondrial Morphology Through Simulation Supervised Learning
12:06

Analyzing Mitochondrial Morphology Through Simulation Supervised Learning

Published on: March 3, 2023

4.6K
Database-guided Flow-cytometry for Evaluation of Bone Marrow Myeloid Cell Maturation
12:05

Database-guided Flow-cytometry for Evaluation of Bone Marrow Myeloid Cell Maturation

Published on: November 3, 2018

12.2K

Area of Science:

  • Hematology and Medical Diagnostics
  • Computational Pathology
  • Artificial Intelligence in Medicine

Background:

  • Bone marrow aspiration and biopsy are standard for diagnosing hematological diseases, but interpretation is subjective and time-consuming.
  • Existing diagnostic methods like flow cytometry (FCM) and molecular analyses have limitations.
  • There is a need for objective, automated systems to analyze bone marrow smears, especially using deep learning.

Purpose of the Study:

  • To develop a deep learning model (BMSNet) to assist hematologists in interpreting bone marrow smears.
  • To enable faster diagnosis and monitoring of hematological diseases.
  • To improve the objectivity and efficiency of bone marrow smear analysis.

Main Methods:

  • Developed BMSNet, a convolutional neural network based on YOLO v3 architecture, for cell detection and classification.
  • Utilized 122 bone marrow smears (2016-2018), divided into development (N=42), validation (N=70), and competition (N=10) cohorts.
  • Eight cell categories were annotated: erythroid, blasts, myeloid, lymphoid, plasma cells, monocyte, megakaryocyte, and unidentified. Human-machine competition with FCM as ground truth.

Main Results:

  • BMSNet demonstrated comparable precision and recall to hematologists in most cell categories after error correction.
  • In detecting >5% blasts, BMSNet's AUC (0.948) exceeded hematologists' (0.929) but was lower than pathologists' (0.985).
  • Performance variations were linked to myelodysplastic syndrome cases; BMSNet showed high correlation (0.960) with FCM in the competition cohort.

Conclusions:

  • The deep learning model BMSNet effectively assists hematologists in bone marrow smear interpretation, accelerating hematopoietic cell detection.
  • BMSNet shows promise for improving diagnostic efficiency and objectivity in hematology.
  • Comprehensive morphological interpretation still necessitates the expertise of trained hematologists.