Predicting Iron Deficiencies Using Routine Complete Blood Cell Count Parameters: A Machine Learning Approach and Evaluation
Vorhersage von Eisenmangel mithilfe routinemäßiger Parameter des kompletten Blutbilds: Machine-Learning-Ansatz und Bewertung
Journal of Clinical Medicine · 19.06.2026 · 15 (12) · 4783
DOI: 10.3390/jcm15124783 ↗Auf einen Blick
In einer monozentrischen retrospektiven Ambulanzstudie sagten CBC-Parameter niedrige Ferritin- und Transferrinsättigungswerte per Machine Learning voraus. Ensemble-Modelle erzielten die beste Leistung.
Fragestellung
Können routinemäßige CBC-Parameter im Machine-Learning-Framework niedrige Ferritin- und Transferrinsättigungswerte vorhersagen?
Methode
Monozentrische retrospektive Ambulanzstudie mit 32.437 Datensätzen; Modelle mit logistischer Regression, Entscheidungsbaum, Random Forest, XGBoost, SVM, k-NN und Naive Bayes; 10-fold Cross-validation und Testset.
Ergebnisse
Alle Modelle zeigten Vorhersagefähigkeit. Die besten AUC-Werte erreichten Random Forest und XGBoost: 0,80–0,87 für Ferritin und 0,85–0,96 für Transferrinsättigung. MCV, MCH und RDW waren wichtige Prädiktoren.
Schlussfolgerung der Autor:innen
CBC-basierte Machine-Learning-Modelle können helfen, Personen mit niedrigem Ferritin oder niedriger Transferrinsättigung zu identifizieren und eine gezielte weitere Abklärung zu unterstützen.
Kernaussagen
- Random Forest und XGBoost erzielten die besten AUC-Werte.
- MCV, MCH und RDW waren Schlüsselprädiktoren.
- Die Vorhersage der Transferrinsättigung war etwas genauer als die des Ferritins.
Deutsche Zusammenfassung: KI-gestützt aus dem Original-Abstract erstellt; alle Zahlen wurden automatisch mit dem Original abgeglichen. Maßgeblich ist der Original-Abstract.
Kurz gefasst
Fachartikel in der Zeitschrift Journal of Clinical Medicine, online erschienen am 19.06.2026. Verfasst von Davide Negrini und 4 weiteren Autor:innen. Beteiligte Einrichtung: Section of Clinical Biochemistry, University of Verona, 37134 Verona, Italy. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background/Objectives: Iron deficiency remains a prevalent condition, needing specific laboratory tests for diagnosis. This study aimed to evaluate whether routine complete blood cell count (CBC) parameters can be used within a machine learning framework to predict low ferritin and low transferrin saturation, used as biochemical markers of altered iron status, potentially supporting more targeted laboratory test utilization.
Methods
In this single-center retrospective outpatient study, we analyzed 32,437 records from subjects undergoing both complete blood cell count and iron metabolism testing between 2023 and 2026. Low ferritin and low transferrin saturation were defined using sex-specific thresholds. Low ferritin was present in 14,344 subjects (44.2%), whereas low transferrin saturation was present in 7791 subjects (24.0%). After cleaning data and excluding incomplete records, demographic variables and CBC indices were tested as potential predictors. The dataset was split into training and test sets with stratified sampling. Multiple supervised machine learning models, including logistic regression, decision tree, random forest, XGBoost, support vector machine, k-nearest neighbors, and Naive Bayes, were trained. Hyperparameter tuning and model selection were performed using repeated stratified 10-fold cross-validation, optimizing the area under the curve (AUC). Model performance was assessed by AUC, sensitivity, and specificity, and validated on an independent test set.
Results
All models showed predictive capability for low ferritin and low transferrin saturation using CBC parameters alone. Ensemble methods, especially random forest and XGBoost, reached the best performance (AUC values of 0.80–0.87 for ferritin and 0.85–0.96 for transferrin saturation). Sensitivity and specificity were balanced, supporting clinical screening applicability. Results were maintained across validation and confirmed in the test set. Prediction of transferrin saturation showed slightly higher accuracy than ferritin. Feature importance analysis identified mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), and red blood cell distribution width (RDW) as key predictors.
Conclusions
CBC-based machine learning models may help identify subjects with low ferritin or low transferrin saturation, supporting subsequent targeted assessment of iron status.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Section of Clinical Biochemistry, University of Verona, 37134 Verona, Italy.
ORCID-Profile: Davide Negrini, Giuseppe Lippi
Zitieren
APA
Negrini, D., Pighi, L., Mignolli, S., Salvagno, G. L., & Lippi, G. (2026). Predicting Iron Deficiencies Using Routine Complete Blood Cell Count Parameters: A Machine Learning Approach and Evaluation. Journal of Clinical Medicine, 15(12), 4783. https://doi.org/10.3390/jcm15124783
Vancouver
Negrini D, Pighi L, Mignolli S, Salvagno GL, Lippi G. Predicting Iron Deficiencies Using Routine Complete Blood Cell Count Parameters: A Machine Learning Approach and Evaluation. Journal of Clinical Medicine. 2026;15(12):4783. doi:10.3390/jcm15124783
BibTeX
@article{Negrini2026,
author = {Negrini, Davide and Pighi, Laura and Mignolli, Simone and Salvagno, Gian Luca and Lippi, Giuseppe},
title = {Predicting Iron Deficiencies Using Routine Complete Blood Cell Count Parameters: A Machine Learning Approach and Evaluation},
journal = {Journal of Clinical Medicine},
year = {2026},
volume = {15},
number = {12},
pages = {4783},
doi = {10.3390/jcm15124783}
}
EUVIMED