A complete blood count-based machine learning model for rapid differentiation of aplastic anemia, immune thrombocytopenia, and myelodysplastic syndromes in routine clinical practice
Practical Laboratory Medicine · 2026-07 · 50 · e00526
DOI: 10.1016/j.plabm.2026.e00526 ↗Kurz gefasst
Fachartikel in der Zeitschrift Practical Laboratory Medicine, 2026-07. Verfasst von Xiaohan Wang und 4 weiteren Autor:innen. Beteiligte Einrichtung: The Second School of Clinical Medicine, Guangdong Medical University, Dongguan, China. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
Background
Accurate differentiation of common hematologic disorders remains challenging in routine clinical practice and often requires invasive diagnostic procedures. Although complete blood count (CBC) testing is widely available, its diagnostic value for early disease triage has not been fully understood.
Methods
Retrospectively, among 165,181 routine blood test records collected between October 2011 and June 2025, 4056 samples with confirmed diagnoses were included for model development and validation after exclusion of cases lacking definitive diagnostic information. Patients were classified into aplastic anemia (AA), immune thrombocytopenia (ITP), myelodysplastic syndrome (MDS), and other hematologic conditions. Machine learning models were developed using routinely available CBC parameters. Model performance was assessed using one-vs-rest receiver operating characteristic (ROC) curves, area under the curve (AUC), and class-specific precision, recall, and F1-scores. Model interpretability was evaluated using Shapley Additive exPlanations (SHAP).
Results
Baseline demographic and hematologic parameters differed significantly among diagnostic groups (all P < 0.001). Among the evaluated models, LightGBM demonstrated robust overall performance, achieving one-vs-rest AUCs of 0.920 for AA, 0.970 for ITP, 0.788 for MDS, and 0.870 for other conditions, with an overall accuracy of 0.82. While AA and ITP were identified with favorable precision and recall, MDS showed lower recall, reflecting substantial overlap in routine laboratory features. SHAP analysis identified platelet count, red blood cell count, and white blood cell count as the most influential predictors.
Conclusions
A machine learning model based on routinely available CBC parameters can support non-invasive differentiation of common hematologic disorders. This approach may serve as a practical screening and triage tool at the outpatient or pre-bone marrow stage, helping optimize the use of invasive diagnostic procedures.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- The Second School of Clinical Medicine, Guangdong Medical University, Dongguan, China.
- State Key Laboratory of Experimental Hematology, National Clinical Research Center for Blood Diseases, Haihe Laboratory of Cell Ecosystem, Institute of Hematology and Blood Diseases Hospital, Chinese Academy of Medical…
- Guangdong Provincial Key Laboratory of Molecular Target & Clinical Pharmacology, the NMPA and State Key Laboratory of Respiratory Disease, the School of Pharmaceutical Sciences, Guangzhou Medical University, Guangzhou,…
- The Sixth Affiliated Hospital of Guangdong Pharmaceutical University, Dongguan, China.
Zitieren
APA
Wang, X., Huang, J., Xi, J., Liu, H., & Yuan, Y. (2026). A complete blood count-based machine learning model for rapid differentiation of aplastic anemia, immune thrombocytopenia, and myelodysplastic syndromes in routine clinical practice. Practical Laboratory Medicine, 50, e00526. https://doi.org/10.1016/j.plabm.2026.e00526
Vancouver
Wang X, Huang J, Xi J, Liu H, Yuan Y. A complete blood count-based machine learning model for rapid differentiation of aplastic anemia, immune thrombocytopenia, and myelodysplastic syndromes in routine clinical practice. Practical Laboratory Medicine. 2026;50:e00526. doi:10.1016/j.plabm.2026.e00526
BibTeX
@article{Wang2026,
author = {Wang, Xiaohan and Huang, Jian and Xi, Jingwen and Liu, Huan and Yuan, Yu},
title = {A complete blood count-based machine learning model for rapid differentiation of aplastic anemia, immune thrombocytopenia, and myelodysplastic syndromes in routine clinical practice},
journal = {Practical Laboratory Medicine},
year = {2026},
volume = {50},
pages = {e00526},
doi = {10.1016/j.plabm.2026.e00526}
}
EUVIMED