A Machine Learning Model Based on CBC-Derived Parameters to Distinguish Benign from Malignant Lymphoproliferative Disorders

Jing Jing, Xiaoyan Hao, Yanjun Diao, Xiang Cheng, Xiaoxia Gao, Bin Huang, Yun Yang, Enliang Hu, Yuan Zhao, Jingyuan Jia, Jiayun LiuResearch Square 2025 · DOI

TL;DR

Distinguishing between benign viral infections like infectious mononucleosis (IM) and malignant blood cancers is notoriously difficult due to overlapping early symptoms. This paper presents a machine learning framework that uses routine and "research-only" parameters from a standard Complete Blood Count (CBC) to accurately differentiate these conditions, achieving an AUC of up to 0.995 with the XGBoost algorithm.

What Problem Does This Paper Address?

In clinical hematology, "look-alike" conditions pose a significant diagnostic challenge. Infectious mononucleosis (IM)—a benign condition—often presents with fever, fatigue, and enlarged lymph nodes, mirroring the early signs of malignant lymphoproliferative disorders like leukemia and lymphoma.

While definitive diagnosis usually requires invasive bone marrow biopsies or expensive flow cytometry, the initial screening tool is almost always the Complete Blood Count (CBC). However, standard CBC reports often provide only a surface-level view. This paper aims to bridge the gap by leveraging "extended" parameters—data points captured by modern hematology analyzers that are usually hidden from the final clinical report—to create a non-invasive, automated triage tool.

Methods

The researchers developed a multicenter diagnostic model using a retrospective cohort. Their approach involved several sophisticated steps:

  • Feature Engineering: They started with 45 parameters (24 routine and 21 research-derived). The research parameters included multidimensional cytometric measurements like "lymphocyte population in side scatter" and "centroid distance between lymphocyte and monocyte populations."
  • Feature Selection: To avoid overfitting and ensure model parsimony, they used an intersection of three methods: Recursive Feature Elimination (RFECV-SVM), Mutual Information (MI), and Spearman correlation. This narrowed the field to 9 key indicators.
  • Model Benchmarking: They trained and validated five different classifiers: XGBoost, AdaBoost, Decision Tree (DT), Logistic Regression (LR), and Gaussian Naïve Bayes (GNB).
  • Explainability: The team employed SHAP (SHapley Additive exPlanations) to interpret the models, moving beyond "black box" predictions to understand which biological features drove the classification.

Datasets

The study utilized data from three independent institutions in Shaanxi Province, China. The total population of 399 participants was categorized into:

  • Infectious Mononucleosis (IM): 141 cases.
  • Hematologic Malignancies: 108 cases (including Acute Lymphoblastic Leukemia, Chronic Lymphocytic Leukemia, and various Lymphomas).
  • Healthy Controls: 150 individuals.

The data was split into a training/validation set (70%) and an independent test set (30%) to ensure the model's generalizability.

Key Insights and Ideas

  • The Power of "Hidden" Data: The inclusion of research parameters (like D_Lym_Ratio_FSC) was vital. These parameters capture the "morphological signature" of cells—their size, internal complexity, and nucleic acid content—which automated analyzers detect but don't usually report.
  • Age as a Biological Proxy: SHAP analysis revealed that age was the most influential variable. This aligns with epidemiological reality: IM is predominantly a disease of the young, while malignant lymphoid disorders are more frequent in older populations.
  • The "Immature Gran" Signal: The presence of immature granulocytes was a strong positive predictor for malignancy, reflecting how cancer disrupts normal blood cell production (myelopoiesis).
  • The Atypical Lymphocyte (Aly) Paradox: Interestingly, higher "Atypical Lymphocyte" flags contributed negatively to a malignancy prediction. In this context, a high Aly count was more indicative of the robust immune response seen in benign IM than in cancer.

Key Results

The XGBoost model emerged as the clear winner, demonstrating superior stability and accuracy:

  • IM vs. Malignancy: In the independent test set, XGBoost achieved an AUC of 0.995, a sensitivity of 92.73%, and an accuracy of 96.80%.
  • Healthy vs. Malignancy: The model maintained high performance with an AUC of 0.934 and an accuracy of 94.62%.
  • Healthy vs. IM: The model was nearly perfect, achieving 100% accuracy in the test set.

Most models (except the Decision Tree) maintained evaluation indicators above 87%, suggesting that the selected 9 features are highly robust diagnostic markers.

Limitations and Future Work

The authors acknowledge several areas for improvement:

  • Geographic Diversity: All data came from three hospitals in a single Chinese province. Future studies need to include more diverse populations to account for regional variations in diet and healthcare access.
  • Sample Size: While sufficient for traditional ML, a larger dataset would allow for more granular classification of specific lymphoma subtypes.
  • Clinical Role: The authors stress that this is a screening and triage tool, not a replacement for gold-standard biopsies. Its primary value lies in "early identification" and "timely intervention," particularly in resource-constrained environments where expert hematomorphologists may not be available.

Summary updated

Abstract

<title>Abstract</title> Objective Infectious mononucleosis (IM) and malignant lymphoproliferative disorders often present with similar initial symptoms and signs. The purpose of study is to develop a machine learning-based model to distinguish IM from lymphoid hematologic malignancies using both routine and research parameters derived from complete blood count (CBC) analysis. Methods The multicenter model development and validation study utilized data from three independent institutions. Patients with a final confirmed diagnosis of infectious mononucleosis (IM), acute lymphoblastic leukemia (ALL), or chronic lymphoproliferative disorders (CLPD) were included. A total of 24 routine and 21 report-derived parameters from the complete blood count (CBC) at initial presentation to our institution were collected. Nine candidate biomarkers and five machine learning classifiers were employed to construct predictive models from the training data. The models were validated using an independent test dataset. Model performance was assessed by calculating the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, accuracy, positive predictive value (PPV), negative predictive value (NPV), Matthews correlation coefficient (MCC), and Cohen’s Kappa coefficient. The diagnostic performance of all five models was evaluated using both internal ROC analysis and external validation datasets. Result A total of 114 patients with infectious mononucleosis (IM), 108 with hematologic diseases, and 150 healthy controls were included in the study. In both the IM versus hematologic disease classification model and the IM versus healthy control model, excluding the decision tree (DT), other methods achieved good evaluation indicators, most of them are above 87% in the validation cohorts, suggesting reliable diagnostic capability. In the classification model distinguishing healthy individuals from patients with hematologic diseases, the XGBoost algorithm showed stable and high performance across the training, validation, and test sets. The receiver operating characteristic (ROC) curves confirmed that XGBoost was the most effective model, the area under the curve (AUC) values was 0.995 in the test sets. Conclusion The XGBoost model demonstrated the most satisfactory performance. Machine learning algorithms show promise for clinical implementation, and the proposed model may aid in the early identification of IM and malignant lymphoid disorders with overlapping initial presentations. This provides valuable assistance for clinical doctors to intervene early and improve prognosis.
BibTeX
@inproceedings{jing2025machine,
  author = {Jing Jing and Xiaoyan Hao and Yanjun Diao and Xiang Cheng and Xiaoxia Gao and Bin Huang and Yun Yang and Enliang Hu and Yuan Zhao and Jingyuan Jia and Jiayun Liu},
  title = {{A Machine Learning Model Based on CBC-Derived Parameters to Distinguish Benign from Malignant Lymphoproliferative Disorders}},
  booktitle = {Research Square},
  year = {2025},
  doi = {10.21203/rs.3.rs-7283529/v1},
}
Figure