DeepHeme, a high-performance, generalizable deep ensemble for bone marrow morphometry and hematologic diagnosis.

Journal: Science translational medicine
Published Date: Jun 11, 2025

Abstract

Cytomorphological analysis of the bone marrow aspirate (BMA) is pivotal for the diagnostic workup of a broad range of hematological disorders. However, this skill is error prone, highly complex, and time consuming. Deep learning-based models for the automatic classification of bone marrow cell morphology demonstrate the potential to improve diagnostic efficiency and accuracy. However, existing deep learning approaches in this field fall short of expert-level performance and lack generalizability beyond a single dataset. Working with multiple hematopathologists, we curated a dataset from the University of California, San Francisco, which included a training set of 30,394 images from 40 patients with morphologically normal marrows and a test set of 8507 images from 10 different patients, all derived from 400×-equivalent whole-slide images (WSIs). We then developed DeepHeme, a snapshot ensemble deep learning classifier, which outperformed previous models in accuracy while expanding the total number of differentiable cell classes. We externally validated DeepHeme using an independent dataset from the Memorial Sloan Kettering Cancer Center, which included 2694 images from 10 morphologically normal patients and 11,076 images from 655 patients with normal or diseased marrow, scanned using a different WSI system, demonstrating robust generalizability. At the level of individual cell classifications, we systematically compared DeepHeme's diagnostic performance with that of three medical experts from different academic hospitals, demonstrating that DeepHeme achieved accuracy comparable to, or exceeding, that of human experts. Accurate and generalizable cell classification represents a step toward automated analysis of hematopathology slides and the development of quantitative, morphology-based, predictive markers.

Authors

  • Shenghuan Sun
    Bakar Computational Health Sciences Institute, University of California, San Francisco, CA 94143, USA.
  • Zhanghan Yin
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Jacob G Van Cleave
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Linlin Wang
    Guangdong-Hong Kong-Macao Greater Bay Area Artificial Intelligence Application Technology Research Institute, Shenzhen Polytechnic University, Shenzhen, China.
  • Brenda Fried
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Khawaja H Bilal
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Fabienne Lucas
    Department of Laboratory Medicine and Pathology, University of Washington, Seattle, Washington, USA.
  • Irem S Isgor
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Dylan C Webb
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Siddharth Singi
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Laura Brown
    Department of Laboratory Medicine, University of California, San Francisco, San Francisco, CA 94143, USA.
  • Roni Shouval
    Roni Shouval, Hila Mishan-Shamay, Avichai Shimoni, and Arnon Nagler, The Chaim Sheba Medical Center, Tel-Hashomer; Roni Shouval, Ori Bondi, and Ron Unger, Bar-Ilan University, Ramat-Gan, Israel; Myriam Labopin, Norbert C. Gorin, Emmanuelle Polge, Arnon Nagler, and Mohamad Mohty, European Group for Blood and Marrow Transplantation; Myriam Labopin and Mohamad Mohty, Sorbonne Universités, Centre de Recherche (CDR) Saint-Antoine; Myriam Labopin and Mohamad Mohty, Institut National de la Santé et de la Recherche Médicale, CDR Saint-Antoine; Myriam Labopin and Mohamad Mohty, Assistance Publique-Hôpitaux de Paris, Hôpital Saint-Antoine, Paris, France; Fabio Ciceri, San Raffaele Scientific Institute, Milan; Andrea Bacigalupo, Ospedale San Martino, Genoa, Italy; Jordi Esteve, Institut d'Investigacions Biomèdiques August Pi i Sunyer, Barcelona, Spain; Sebastian Giebel, Maria Sklodowska-Curie Memorial Cancer Center and Institute of Oncology, Gliwice, Poland; Christoph Schmid, Ludwig-Maximilians-University, Munich; Nicolaus Kroger, University Medical Center Hamburg Eppendorf, Hamburg, Germany; Mahmoud Aljurf, King Faisal Specialist Hospital & Research Centre, Riyadh, Saudi Arabia; Charles Craddock, Queen Elizabeth Hospital, Birmingham, United Kingdom; Jan J. Cornelissen, Erasmus University Medical Center, Rotterdam, the Netherlands; and Frederic Baron, University of Liège, Liège, Belgium. shouval@gmail.com.
  • Jeff Lin
    Department of Bioengineering and Therapeutic Sciences, University of California, San Francisco, San Francisco, CA 94143, USA.
  • Ethan S Yan
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Jacob D Spector
    Department of Laboratory Medicine, University of California, San Francisco, San Francisco, CA, USA.
  • Orly Ardon
    ARUP Laboratories, Salt Lake City, Utah, USA.
  • Leonardo Boiocchi
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Rohan Sardana
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Jeeyeon Baik
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Menglei Zhu
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Aijazuddin Syed
    Department of Pathology, Memorial Sloan Kettering Cancer Center, New York, New York.
  • Mariko Yabe
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Chuanyi M Lu
    Department of Laboratory Medicine, University of California, San Francisco, San Francisco, CA 94143, USA.
  • Mikhail Roshal
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Chad Vanderbilt
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
  • Dmitry B Goldgof
    Department of Computer Science, University of South Florida, Tampa, FL 33620, USA.
  • Ahmet Dogan
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Sonam Prakash
    Department of Laboratory Medicine, University of California, San Francisco, San Francisco, CA 94143, USA.
  • Iain Carmichael
    Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY 10065, USA.
  • Atul J Butte
    Bakar Computational Health Sciences Institute, University of California San Francisco, San Francisco, CA.
  • Gregory M Goldgof
    Bakar Computational Health Sciences Institute, University of California, San Francisco, CA 94143, USA.

Keywords

External Resources

Popular Topics
Recent Journals