Predictors of Anemia in Ethiopia: A Systematic-Review of Machine Learning Approaches
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Anemia remains a critical public health issue globally which is disproportionately affecting population in low- and middle income countries, with sub-Saharan Africa particularly Ethiopia experiencing high prevalence rates, Despite ongoing interventions, understanding the multifactorial causes of anemia and enhancing predictive Modelling through modern analytic approaches remains limited.
Objective
This systematic review aims to evaluate and synthesize current evidence on the application of machine learning algorithm for predicting anemia among various population in Ethiopia, focusing on identifying predictive models used, key predictors and methodological strength of those existing studies.
Methods
Following the PRISMA 2020 guidelines, a comprehensive search was conducted across PubMed, science Direct, HINARI, and Google Scholar from October 25 to November 10 2024. Observational studies employing Machine learning algorithm to predict anemia in Ethiopia were included. The quality of included studies was evaluated using BSA Medical Sociology Group Assessment Tool.
Results
Out of 513 initially retrieved records, four studies met the inclusion criteria. These Studies targeted Children under five, pregnant women, and young girls, Utilizing algorithms such as Random Forest, Logistic Regression and Boruta algorithm .the random Forest model emerged as the most frequently and effectively used technique due to its Robustness and capacity for handling complex data. Anemia prevalence across the included studies ranged from 26% to 57%. A total of 28 unique predictor variables were identified.
Conclusion and Recommendation
Machine learning algorithms, particularly random forest, offer promising tools for accurately predicting anemia in Ethiopia by integrating wide range of socio-demographic and health-related factors, however, the limited number of studies and population specific focus highlight the need for more comprehensive and generalizable research to inform effective public health intervention.
