System Maintenance Notice

Please note that maintenance work is scheduled between 4 AM and 5 AM - CET. During this time, you may experience reduced performance. Thanks for your patience.

Publications

Journal Article Ppublish

PubMed: 42242261 DOI: 10.1016/S2352-3026(26)00099-2

Prediction of bone marrow fibrosis from complete blood count in myeloproliferative neoplasms (FIBOM-AI): a multicentre machine learning model development and validation study.

Lancet Haematol 13(6):e396-e407

Authors: Marie Donzel, Syrine Khedimallah, Sabrina Barriere, Agathe Gouband, Christopher Aubert, Nelly Bosselut, Elsa Poullot, Silvia Maria Bezsera, Olivier Mansier, Nursemin Yilmaz, Charlotte Syrykh, Xavier Hittinger, Seddig Momenkhan, Fiorenza Barraco, Lila Gilis, Hélène Labussière-Wallet, Mathias Brehon, Benjamin Chin-Yee, Cyrus Hsia, Luojun Wang, Sarah Zouaghi, Pierre Sujobert, Marion Eveillard, Véronique De Mas, Suzanne Tavitian, Laurence Legros, Marie Parrens, Pascale Flandrin-Gresta, Mathieu Meunier, Nicolas Freynet, Lydia Roy, Véronique Meignin, Louis Drevon, Dana Ranta, Simona Pagliuca, Eric Lippert, Jean-Christophe Ianotto, Laurent Jallades, Barbara Burroni, Damien Luque Paz, Alexandra Traverse-Glehen, Simon Chevalier, Vincent Alcazer

Affiliations

  • Service d'anatomopathologie, Centre Hospitalier Lyon Sud, Hospices Civils de Lyon, Pierre Bénite, France.
  • Laboratoire d'hématologie, CHU de Grenoble, Grenoble, France.
  • Service d'hématologie Clinique, CHU de Clermont-Ferrand, Clermont-Ferrand, France.
  • Service d'hématologie Clinique, CHU d'Angers, Angers, France.
  • Service d'hématologie, CHU de Nancy, France.
  • Laboratoire d'hématologie, Hôpital Saint-Louis, AP-HP, Paris, France.
  • Département d'anatomopathologie, Hôpitaux Universitaires Henri Mondor, AP-HP, Créteil, France.
  • Service d'hématologie, CHU de Saint Etienne, Saint Etienne, France.
  • Laboratoire d'Hématologie, Centre Hospitalier Universitaire de Bordeaux, Bordeaux, France.
  • Service d'hématologie Clinique, Hôpital Bicêtre, AP-HP, Le Kremlin-Bicêtre, France.
  • Département d'anatomopathologie, Institut Universitaire du Cancer-Oncopole de Toulouse, Toulouse, France.
  • Hematology Biology, CHU Nantes, Nantes Université, INSERM, CNRS, CRCI2NA, Nantes, France.
  • Service d'hématologie Clinique, Centre Hospitalier Lyon Sud, Hospices Civils de Lyon, Pierre Bénite, France.
  • Department of Pathology and Laboratory Medicine, London Health Sciences Centre, London, Canada.
  • Laboratoire d'hématologie, Centre Hospitalier Lyon Sud, Hospices Civils de Lyon, Pierre Bénite, France.
  • Service d'anatomie et de cytologie pathologiques, CHU de Bordeaux, & Inserm, UMR1312, Bordeaux Institute of Oncology, Université de Bordeaux, Bordeaux, France.
  • Université Grenoble Alpes, INSERM U1209, CNRS UMR 5309, CHU Grenoble Alpes, Institut pour l'Avancée des Biosciences, 38000 GRENOBLE, France.
  • Département d'hématologie biologique et d'immunologie, Hôpitaux Universitaires Henri Mondor, AP-HP, Créteil, France.
  • Service d'hématologie Clinique, Hôpitaux Universitaires Henri Mondor, AP-HP, Université Paris Est Créteil, Créteil, France.
  • Service d'Anatomie et Cytologie Pathologiques, Hôpital Saint-Louis, AP-HP, Paris, France.
  • Laboratoire d'hématologie, CHU de Brest, Brest, France.
  • Service d'Hématologie et d'Hémostase clinique, CHU de Brest, Brest, France.
  • Service de pathologie, Hôpital Cochin, AP-HP, Paris, France.
  • Service d'hématologie Clinique, Centre Hospitalier Lyon Sud, Hospices Civils de Lyon, Pierre Bénite, France; Centre International de Recherche en Infectiologie, INSERM U1111-CNRS UMR5308, Université Lyon 1, Lyon, France.

Abstract

BACKGROUND

Bone marrow biopsy is essential for myeloproliferative neoplasms diagnosis, particularly for assessing bone marrow fibrosis. However, it is an invasive procedure that can be difficult to perform in some patients. This study aimed to develop an artificial intelligence (AI) model (FIBOM-AI) to predict grade 2-3 fibrosis using complete blood count (CBC) data and age.

METHODS

In this machine learning model, histopathological and CBC data from patients (no age restriction) undergoing bone marrow biopsies, performed either at diagnosis or during follow-up for suspected or established myeloproliferative neoplasm, were retrospectively collected across 13 French university hospitals. 27 CBC variables and patient age at biopsy were used as predictors. Data from six independent centres were used for model development and the remaining seven centres were used for external validation. The primary outcome was grade 2-3 fibrosis as a binary endpoint. Nine algorithms were trained using 10-fold cross validation. The final model was evaluated in a real-life prospective setting on all bone marrow biopsies performed in four French centres and one Canadian centre. To guide clinical decisions, two different predictions based on the final model were set: an overall prediction, maximising the overall accuracy of the model, and a confident prediction, maximising either sensitivity (rule-out) or specificity (rule-in).

FINDINGS

Between Jan 1, 2014, and Dec 31, 2023, 1995 patients from 13 centres were included in the retrospective cohort; the median age was 62·0 years (IQR 50·0-71·0), 1062 (53·2%) were male, and 933 (46·8%) were female. Between Jan 1, 2024, and Dec 31, 2024, 493 patients from five centres were included in the prospective cohort; the median age was 65·0 years (IQR 53·6-73·0), 297 (60·2%) were male, and 196 (39·8%) were female. The Extreme Gradient Boosting model had the best performance, with an area under the curve of 0·96 (95% CI 0·95-0·97), 0·90 (0·85-0·95), and 0·92 (0·90-0·95) on the training, testing, and validation sets, respectively. The final accuracy was 88·1%, 87·0%, and 88·6% for the overall predictions and 99·5%, 94·5%, and 96·9% for the confident predictions in the training, testing, and validation sets, respectively. Prospective evaluation showed an accuracy of 85·2% (range 82·0-87·3) for the overall predictions and 98·6% (96·0-100·0) for the confident predictions.

INTERPRETATION

FIBOM-AI can be reliably used as a triage and prioritisation tool to support timely bone marrow evaluation or provide complementary risk assessment when bone marrow biopsy is not immediately feasible.

FUNDING

None.

MeSH

Research Design, Bone Marrow, Blood, Fibrosis, Neoplasms, Biopsy, Machine Learning, Diagnosis, Learning, Artificial Intelligence, Primary Myelofibrosis