Artificial intelligence-based quantitative bone marrow pathology analysis for myeloproliferative neoplasms
Authors:
- Dandan YuDYDandan Yu
- Rongfeng FuRFRongfeng Fu
- Ting SunTSTing Sun
- Huan DongHDHuan Dong
- Wenjing GuWGWenjing Gu
- Renchi YangRYRenchi Yang
- Zhijian XiaoZXZhijian Xiao
Open Access
- 12 June 2025
- journal article
- Published by Ferrata Storti Foundation (Haematologica) in Haematologica
- Vol. 110 (11) , 2691-2701
Abstract
The evaluation of bone marrow pathology is essential for diagnosing and classifying myeloproliferative neoplasms (MPNs). However, morphological assessments of bone marrow trephine (BMT) sections by hematopathologists are inherently subjective; thus, an accurate and objective diagnostic system is needed. Based on U2-Net, UNeXt, and ResNet, we developed an automatic quantitative analysis platform of BMT sections from MPNs patients and nonneoplastic cases (n=342 total) to enhance the accuracy of diagnosis and classification of MPNs. Bone marrow metrics, including marrow cellularity, the myeloid-to-erythroid (M: E) ratio, megakaryocyte morphology and distribution, and the grading of marrow fibrosis (MF), were quantitatively analyzed (with an accuracy of approximately 0.9) based on the accuracy segmentation and identification of various cells and tissues (with an intersection over union (IoU) of roughly 0.8). A bone marrow classification model incorporating bone marrow metrics, a clinical classification model utilizing clinical features, and a comprehensive classification model that includes both bone marrow metrics and clinical features were developed using random forest classifiers to differentiate MPN subtypes and nonneoplastic conditions. The bone marrow and comprehensive classification models reached a macro-average area under the curve (AUC) of 0.96 for differentiating MPN subtypes and nonneoplastic cases. The clinical classification model attained a macro-average AUC of 0.92. This platform is highly accurate for quantitatively analyzing bone marrow pathology and classifying MPN subtypes and nonneoplastic cases. It can be a potentially auxiliary diagnostic tool for hematopathologists when dealing with patients with suspected MPNs.Keywords
This publication has 32 references indexed in Scilit:
- 3023 Mayo Clinic Patients With Myeloproliferative Neoplasms: Risk-Stratified Comparison of Survival and Outcomes Data Among Disease SubgroupsMayo Clinic Proceedings, 2019
- The 2016 WHO classification and diagnostic criteria for myeloproliferative neoplasms: document summary and in-depth discussionBlood Cancer Journal, 2018
- An application of machine learning to haematological diagnosisScientific Reports, 2018
- Clinical course and outcome of essential thrombocythemia and prefibrotic myelofibrosis according to the revised WHO 2016 diagnostic criteriaOncotarget, 2017
- Presentation and outcome of patients with 2016 WHO diagnosis of prefibrotic and overt primary myelofibrosisBlood, 2017
- Emerging Themes in Image Informatics and Molecular Analysis for Digital PathologyAnnual Review of Biomedical Engineering, 2016
- The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemiaBlood, 2016
- WHO‐histological criteria for myeloproliferative neoplasms: reproducibility, diagnostic accuracy and correlation with gene mutations and clinical outcomesBritish Journal of Haematology, 2014
- Survival and Disease Progression in Essential Thrombocythemia Are Significantly Influenced by Accurate Morphologic Diagnosis: An International StudyJournal of Clinical Oncology, 2011
- Digital images and the future of digital pathologyJournal of Pathology Informatics, 2010