Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: May 31, 2026

A Point-of-Care Method with Integrated Decision Support Tool to Estimate Anemia at Population Level
05:35

A Point-of-Care Method with Integrated Decision Support Tool to Estimate Anemia at Population Level

Published on: January 19, 2024

Beyond prediction: explainable machine learning for ferritin estimation from complete blood count.

Ramazan Berk Us1, Oguzhan Panatli2,3, Mine Büşra Bozkürk4

  • 1Department of Medical Biochemistry, Ankara Etlik City Hospital, Yenimahalle, Ankara, 06170, Türkiye. berkus62@gmail.com.

BMC Medical Informatics and Decision Making
|May 29, 2026
PubMed
Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

The grand piano ratio: an adjunctive intraoperative screening tool for detecting excessive femoral component external rotation in total knee arthroplasty.

Archives of orthopaedic and trauma surgery·2026
Same author

Comparing conventional correction formulas and machine learning-based prediction of ionized calcium.

Clinica chimica acta; international journal of clinical chemistry·2026
Same author

Inferior Screw Referenced Calcar Tip Apex Distance as the Most Accurate Predictor of Mechanical Cut Out in Dual-Screw Proximal Femoral Nails.

Medicina (Kaunas, Lithuania)·2026
Same author

Serum paraoxonase-1 as a marker of oxidative stress and pulmonary dysfunction in sarcoidosis: association with disease activity and prognostic potential.

Frontiers in immunology·2026
Same author

May serum advanced oxidation protein products levels be diagnostic markers in endometrioma?

BMC women's health·2025
Same author

Lower ANXA3 Levels May Be Related to Major Depressive Disorder.

Life (Basel, Switzerland)·2025

Machine learning models using complete blood count (CBC) data can reliably screen for iron deficiency without costly ferritin tests. Explainable AI enhances these models, offering a zero-cost clinical decision support tool for iron status assessment.

Area of Science:

  • Biomedical Informatics
  • Hematology
  • Machine Learning

Background:

  • Serum ferritin is a key iron marker but is expensive and elevated during inflammation.
  • Complete blood count (CBC) parameters are widely available and cost-effective.
  • Developing alternative methods to assess iron status is crucial for accessible healthcare.

Purpose of the Study:

  • To develop and validate explainable machine learning (ML) models for estimating ferritin status.
  • To utilize only complete blood count (CBC) and demographic data for iron status prediction.
  • To provide a cost-effective and reliable screening tool for iron deficiency.

Main Methods:

  • Prospective analysis of 12,450 adult laboratory records, excluding inflammatory cases.
  • Training five supervised ML algorithms (Logistic Regression, XGBoost, ANN) on a balanced dataset of 7,200 samples.
Keywords:
Clinical decision support systemsComplete blood countExplainable artificial intelligenceFerritinIron deficiencySHAP

Related Experiment Videos

Last Updated: May 31, 2026

A Point-of-Care Method with Integrated Decision Support Tool to Estimate Anemia at Population Level
05:35

A Point-of-Care Method with Integrated Decision Support Tool to Estimate Anemia at Population Level

Published on: January 19, 2024

  • Utilizing SHapley Additive Explanations (SHAP) for model interpretability and feature importance assessment.

    • Main Results:

    • The Artificial Neural Network (ANN) model showed strong classification performance (AUC 0.837, F1-score 0.760) for low ferritin levels.
    • Regression models predicted ferritin levels with higher accuracy in critical low ranges.
    • SHAP analysis identified Sex, Mean Corpuscular Hemoglobin, and Red Cell Distribution Width as key predictors.

    Conclusions:

    • Machine learning models using routine hemogram data offer a reliable, zero-cost screening for iron deficiency.
    • Explainable AI integration increases clinical trust and facilitates implementation.
    • These models serve as an accessible decision support tool for iron status assessment.