跳到正文
原文
Nature 健康护理主题·· 2 天前AI 评分35

使用可解释的集成机器学习模型预测埃塞俄比亚孕妇贫血

Prediction of anemia among pregnant women in Ethiopia using explainable ensemble machine learning models

AI 导读

该研究利用埃塞俄比亚人口代表性调查数据,开发并评估了集成机器学习模型以预测孕妇贫血。Gradient Boosting模型表现出最高的区分能力,ROC-AUC值为0.65。研究发现媒体接触史、流产史、妊娠意图、宗教信仰和铁补充剂使用是影响孕妇贫血的关键因素。该模型在产前护理中具有一定的预测潜力,但需提高敏感度、纳入纵向和临床数据并进行外部验证。

正文

We’re sharing this article early to provide faster access to peer-reviewed, accepted research. It is citable and carries a permanent DOI. This version is subject to further edits and will be replaced automatically by the final Version of Record. All legal disclaimers apply.

Abstract

Anemia during pregnancy remains a major public health concern in Ethiopia, contributing significantly to maternal and fetal morbidity and mortality. This study aimed to develop and evaluate an ensemble machine learning–based model for predicting anemia among pregnant women in Ethiopia using nationally representative survey data. A total of 2699 pregnant women with complete hemoglobin measurements and covariate information were included from the Ethiopian Demographic and Health Surveys (EDHS) of 2005, 2011, and 2016. Six machine learning models (Logistic Regression, Random Forest, Support Vector Machine, Gradient Boosting, CatBoost, and stacking ensemble) were trained and evaluated using performance evaluation metrics. Gradient Boosting demonstrated the highest overall discriminatory ability, achieving a ROC-AUC of 0.65, followed closely by CatBoost and the stacking ensemble model. SHapley Additive exPlanations (SHAP) analysis revealed that media exposure, history of pregnancy termination, pregnancy intention, religion, and iron supplementation were among the most influential predictors of anemia among pregnant women. An ensemble machine learning model showed moderate potential for predicting anemia among pregnant women in Ethiopia and may serve as a supportive decision-making tool in antenatal care settings. However, improvements in sensitivity, incorporation of longitudinal and clinical data, and external validation are required.

Abbreviations

ANN:

Artificial neural network

ANC:

Antenatal care

AUC:

Area under the curve

CI:

Confidence interval

DHS:

Demographic and health survey

GBM:

Gradient Boosting Machine

HGB:

Hemoglobin

LR:

Logistic regression

ML:

Machine learning

NPV:

Negative predictive value

PPV:

Positive predictive value

RF:

Random forest

RFE:

Recursive feature elimination

ROC:

Receiver operating characteristic

SHAP:

SHapley additive exPlanations

SMOTE:

Synthetic minority over–sampling technique

SVM:

Support vector machine

WHO:

World health organization

Acknowledgements

We would like to acknowledge DHS program of the United States Government for providing the authorization to access the Ethiopian DHS of 2005, 2011, and 2016.

Funding

There is no specific funding available.

Author information

Authors and Affiliations

  1. Department of Medical Laboratory Science, College of Medicine and Health Sciences, Debre Markos University, Debre Markos, Ethiopia

    Aytenew Atnaf, Desalegn Abebaw & Sefefe Birhanu Tizie

  2. Department of Health Informatics, College of Medicine and Health Sciences, Debre Markos University, Debre Markos, Ethiopia

    Aytenew Atnaf, Mulatu Gebeyehu, Zigale Yalew Getie, Yohanis Mengist Amare, Sewnet Beyene, Muluken Belachew Mengistie, Abraham Keffale Mengistu, Sefefe Birhanu Tizie, Ashagrie Anteneh & Maru Meseret Tadele

Authors

  1. Aytenew Atnaf
  2. Desalegn Abebaw
  3. Mulatu Gebeyehu
  4. Zigale Yalew Getie
  5. Yohanis Mengist Amare
  6. Sewnet Beyene
  7. Muluken Belachew Mengistie
  8. Abraham Keffale Mengistu
  9. Sefefe Birhanu Tizie
  10. Ashagrie Anteneh
  11. Maru Meseret Tadele

Corresponding author

Correspondence to Aytenew Atnaf.

Ethics declarations

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

The data source for this secondary data was obtained from DHS. The consent to access these datasets was approved by Institutional Review Board (IRB) of DHS program on 15th December, 2025 with reference webpage https://dhsprogram.com/Data/terms-of-use.cfm.

Consent for publication

All authors declared that they have no conflict of interest for publication.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Atnaf, A., Abebaw, D., Gebeyehu, M. et al. Prediction of anemia among pregnant women in Ethiopia using explainable ensemble machine learning models. Sci Rep (2026). https://doi.org/10.1038/s41598-026-81082-z

Download citation

  • Received: 05 August 2026

  • Accepted: 12 November 2026

  • Published: 18 November 2026

  • DOI: https://doi.org/10.1038/s41598-026-81082-z

Keywords

来源:Nature 健康护理主题 · nature.com