EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Prediction of B/T Subtype and ETV6–RUNX1 Translocation in Pediatric Acute Lymphoblastic Leukemia by Deep Learning Analysis of Giemsa‐Stained Whole Slide Images of Bone Marrow Aspirates

Arkadi Piven iD, Gil Shamai iD, Sarah Elitzur iD, Galit Pinto Berger, Yoav Binenbaum iD, Ron Kimmel iD, Ronit Elhasid iD

Pediatric Blood & Cancer · 2025

DOI: 10.1002/pbc.31797 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Pediatric Blood & Cancer, 2025. Verfasst von Arkadi Piven und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science, Technion Israel Institute of Technology, Haifa, Israel. Hauptthemen laut MEDLINE: Deep Learning, Oncogene Proteins, Fusion, Core Binding Factor Alpha 2 Subunit, Translocation, Genetic. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 40 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Israel Science Foundation (ISF), Israel Science Foundation (ISF) - Israel Precision Medicine Partnership (IPMP) program, Israel Innovation Authority (Kamin program).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Accurate determination of B/T‐cell lineage and the presence of the ETV6–RUNX1 translocation is critical for diagnosing acute lymphoblastic leukemia (ALL), as these factors influence treatment decisions and outcomes. However, these diagnostic processes often rely on advanced tools unavailable in low‐resource settings, creating a need for alternative solutions. Procedure We developed a deep learning pipeline to analyze Giemsa‐stained bone marrow (BM) aspirate smears. The models were trained to distinguish between ALL, acute myeloid leukemia (AML), and non‐leukemic BM samples, predict B‐ and T‐cell lineage in ALL, and detect the presence of the ETV6–RUNX1 translocation. The performance was evaluated using cross‐validation (CV) and an external validation cohort.

Results

The models achieved a statistically significant area under the curve (AUC) of 0.99 in distinguishing ALL from AML and control samples. In cross‐validation (CV), the models achieved a cross‐validation AUC of 0.74 for predicting B/T subtypes. For predicting ETV6–RUNX1 translocation, the models achieved an AUC of 0.80. External cohort validation confirmed significant AUCs of 0.72 for B/T subtype classification and 0.69 for ETV6–RUNX1 translocation prediction.

Conclusions

Convolutional neural networks (CNNs) demonstrate potential as a diagnostic tool for pediatric ALL, enabling the identification of B/T lineage and ETV6–RUNX1 translocation from Giemsa‐stained smears. These results pave the way for future utilization of CNNs as a diagnostic modality for pediatric leukemia in low‐resource settings, where access to advanced diagnostic techniques is limited.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningOncogene Proteins, FusionCore Binding Factor Alpha 2 SubunitTranslocation, GeneticPrecursor Cell Lymphoblastic Leukemia-LymphomaBone MarrowHumansChildMaleChild, PreschoolFemaleAdolescentInfantPrognosisETS Translocation Variant 6 Protein

Stichwörter der Autor:innen

B/T lineage classificationETV6–RUNX1 translocationdeep learninggiemsa‐stained bone marrow smearspediatric acute lymphoblastic leukemia

Substanzen

Oncogene Proteins, FusionCore Binding Factor Alpha 2 SubunitTEL-AML1 fusion proteinETS Translocation Variant 6 Protein

Autor:innen und Institutionen

  • Department of Computer Science, Technion Israel Institute of Technology, Haifa, Israel.
  • Department of Hematology and Oncology, Schneider Children's Medical Center of Israel, Petah Tikva, Israel.
  • Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel.
  • Department of Oncology, St. Jude Children's Hospital, Memphis, Tennessee, USA.
  • Division of Pediatric Hematology-Oncology, Boston Children's Hospital, Boston, Massachusetts, USA.
  • Department of Pediatric Oncology, Dana-Farber Cancer Institute, Boston, Massachusetts, USA.

ORCID-Profile: Arkadi Piven, Gil Shamai, Sarah Elitzur, Yoav Binenbaum, Ron Kimmel, Ronit Elhasid

Förderung

  • Israel Science Foundation (ISF)
  • Israel Science Foundation (ISF) - Israel Precision Medicine Partnership (IPMP) program
  • Israel Innovation Authority (Kamin program)

Zitieren

APA

Piven, A., Shamai, G., Elitzur, S., Berger, G. P., Binenbaum, Y., Kimmel, R., & Elhasid, R. (2025). Prediction of B/T Subtype and ETV6–RUNX1 Translocation in Pediatric Acute Lymphoblastic Leukemia by Deep Learning Analysis of Giemsa‐Stained Whole Slide Images of Bone Marrow Aspirates. Pediatric Blood & Cancer. https://doi.org/10.1002/pbc.31797

Vancouver

Piven A, Shamai G, Elitzur S, Berger GP, Binenbaum Y, Kimmel R, et al. Prediction of B/T Subtype and ETV6–RUNX1 Translocation in Pediatric Acute Lymphoblastic Leukemia by Deep Learning Analysis of Giemsa‐Stained Whole Slide Images of Bone Marrow Aspirates. Pediatric Blood & Cancer. 2025. doi:10.1002/pbc.31797

BibTeX
@article{Piven2025,
  author = {Piven, Arkadi and Shamai, Gil and Elitzur, Sarah and Berger, Galit Pinto and Binenbaum, Yoav and Kimmel, Ron and Elhasid, Ronit},
  title = {Prediction of B/T Subtype and ETV6–RUNX1 Translocation in Pediatric Acute Lymphoblastic Leukemia by Deep Learning Analysis of Giemsa‐Stained Whole Slide Images of Bone Marrow Aspirates},
  journal = {Pediatric Blood & Cancer},
  year = {2025},
  doi = {10.1002/pbc.31797}
}
Kostenlos starten →