Applying Machine learning to analyze the utilization of insecticide-treated nets among rural under-five children in East Africa

IF 2.4 Q3 INFECTIOUS DISEASES

Parasite Epidemiology and Control Pub Date : 2026-05-01 Epub Date: 2026-05-12 DOI:10.1016/j.parepi.2026.e00511

Fentahun Bikale Kebede , Amanuel Worku , Angwach Abrham Asnake , Eliyas Addisu Taye , Kussie Urmale Mare , Abiy Mergia Bulto , Abraham Keffale Mengistu , Tsegereda Abebe Andargie , Bewuketu Terefe

{"title":"Applying Machine learning to analyze the utilization of insecticide-treated nets among rural under-five children in East Africa","authors":"Fentahun Bikale Kebede , Amanuel Worku , Angwach Abrham Asnake , Eliyas Addisu Taye , Kussie Urmale Mare , Abiy Mergia Bulto , Abraham Keffale Mengistu , Tsegereda Abebe Andargie , Bewuketu Terefe","doi":"10.1016/j.parepi.2026.e00511","DOIUrl":null,"url":null,"abstract":"<div><h3>Introduction</h3><div>Malaria continues to be a major cause of morbidity and mortality in children under the age of five in the sub-Saharan region of Africa. Despite being one of the pillars of prevention, the use of Insecticide-Treated Nets (ITNs) is still low in rural East Africa. This research aimed to forecast the use of ITNs by rural communities with children in this region using machine learning (ML) and identify the most predictive factors.</div></div><div><h3>Methods</h3><div>We analyzed pooled data from the Demographic and Health Surveys (DHS) in 11 East African nations between 2011 and 2022. Six ML classification models, K-Nearest Neighbours, Support Vector Classifier (SVC), Random Forest, Multilayer Perceptron (MLP), Naïve Bayes, and Logistic Regression, were developed to predict ITN adoption. Model performance was evaluated using Accuracy, Precision, Recall, F1-Score, and ROC-AUC. Shapley Additive exPlanations (SHAP) analysis was used to improve model interpretability and to rank feature importance.</div></div><div><h3>Result</h3><div>In 98,684 weighted household samples, the prevalence of general ITN use was 47.12%. The MLP model generalized best overall, achieving the highest accuracy (91.56%) and ROC-AUC (97.15%), with very high precision (87.95%) and recall (95.24%). While the SVC model had better recall (99.92%), its precision (84.19%) was significantly lower, indicating a higher proportion of false positives. MLP's F1-score of 91.39% restored it as the most balanced classifier in this scenario. SHAP analysis identified the number of children who slept under ITNs and the number of ITNs in the home as the strongest predictors. More community-level wealth, household wealth, and media exposure were also positively associated with ITN use.</div></div><div><h3>Conclusion</h3><div>Machine learning made accurate predictions of ITN use, with the MLP model having the best and most balanced performance. SHAP allowed us to uncover actionable reasons for the most influential drivers of ITN use. These evidence-based outcomes can inform targeted public health interventions to improve ITN coverage and utilization in high-risk rural settings, ultimately reducing malaria cases and deaths.</div></div>","PeriodicalId":37873,"journal":{"name":"Parasite Epidemiology and Control","volume":"33 ","pages":"Article e00511"},"PeriodicalIF":2.4000,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Parasite Epidemiology and Control","FirstCategoryId":"1085","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S240567312600036X","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2026/5/12 0:00:00","PubModel":"Epub","JCR":"Q3","JCRName":"INFECTIOUS DISEASES","Score":null,"Total":0}

引用次数: 0

Abstract

Introduction

Malaria continues to be a major cause of morbidity and mortality in children under the age of five in the sub-Saharan region of Africa. Despite being one of the pillars of prevention, the use of Insecticide-Treated Nets (ITNs) is still low in rural East Africa. This research aimed to forecast the use of ITNs by rural communities with children in this region using machine learning (ML) and identify the most predictive factors.

Methods

We analyzed pooled data from the Demographic and Health Surveys (DHS) in 11 East African nations between 2011 and 2022. Six ML classification models, K-Nearest Neighbours, Support Vector Classifier (SVC), Random Forest, Multilayer Perceptron (MLP), Naïve Bayes, and Logistic Regression, were developed to predict ITN adoption. Model performance was evaluated using Accuracy, Precision, Recall, F1-Score, and ROC-AUC. Shapley Additive exPlanations (SHAP) analysis was used to improve model interpretability and to rank feature importance.

Result

In 98,684 weighted household samples, the prevalence of general ITN use was 47.12%. The MLP model generalized best overall, achieving the highest accuracy (91.56%) and ROC-AUC (97.15%), with very high precision (87.95%) and recall (95.24%). While the SVC model had better recall (99.92%), its precision (84.19%) was significantly lower, indicating a higher proportion of false positives. MLP's F1-score of 91.39% restored it as the most balanced classifier in this scenario. SHAP analysis identified the number of children who slept under ITNs and the number of ITNs in the home as the strongest predictors. More community-level wealth, household wealth, and media exposure were also positively associated with ITN use.

Conclusion

Machine learning made accurate predictions of ITN use, with the MLP model having the best and most balanced performance. SHAP allowed us to uncover actionable reasons for the most influential drivers of ITN use. These evidence-based outcomes can inform targeted public health interventions to improve ITN coverage and utilization in high-risk rural settings, ultimately reducing malaria cases and deaths.

查看原文本刊更多论文

应用机器学习分析东非农村五岁以下儿童使用驱虫蚊帐的情况

疟疾仍然是撒哈拉以南非洲地区5岁以下儿童发病和死亡的一个主要原因。尽管驱虫蚊帐是预防的支柱之一，但在东非农村，驱虫蚊帐的使用率仍然很低。本研究旨在利用机器学习（ML）预测该地区有儿童的农村社区对itn的使用情况，并确定最具预测性的因素。方法：我们分析了2011年至2022年11个东非国家人口与健康调查（DHS）的汇总数据。六种ML分类模型，k近邻，支持向量分类器（SVC），随机森林，多层感知器（MLP）， Naïve贝叶斯和逻辑回归，被开发用于预测ITN的采用。模型性能评估使用准确性、精度、召回率、F1-Score和ROC-AUC。Shapley加性解释（SHAP）分析用于提高模型的可解释性并对特征重要性进行排序。结果98684份加权家庭样本中，一般ITN使用率为47.12%；MLP模型总体泛化效果最好，准确率最高（91.56%），ROC-AUC最高（97.15%），准确率（87.95%）和召回率（95.24%）非常高。虽然SVC模型具有更好的召回率（99.92%），但其准确率（84.19%）明显较低，表明假阳性比例较高。MLP的f1得分为91.39%，使其恢复为该场景下最平衡的分类器。SHAP分析发现，在ITNs下睡觉的儿童数量和家中ITNs的数量是最强的预测因子。更多的社区财富、家庭财富和媒体曝光也与ITN的使用呈正相关。结论机器学习能准确预测ITN的使用情况，其中MLP模型的预测效果最好、最平衡。SHAP使我们能够发现最具影响力的ITN使用驱动因素的可操作原因。这些循证结果可为有针对性的公共卫生干预措施提供信息，以改善高风险农村环境中ITN的覆盖和利用，最终减少疟疾病例和死亡。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

Parasite Epidemiology and Control Medicine-Infectious Diseases

CiteScore

5.70

自引率

3.10%

发文量

审稿时长

17 weeks

期刊介绍： Parasite Epidemiology and Control is an Open Access journal. There is an increasing amount of research in the parasitology area that analyses the patterns, causes, and effects of health and disease conditions in defined populations. This epidemiology of parasite infectious diseases is predominantly studied in human populations but also spans other major hosts of parasitic infections and as such this journal will have a broad remit. We will focus on the major areas of epidemiological study including disease etiology, disease surveillance, drug resistance and geographical spread and screening, biomonitoring, and comparisons of treatment effects in clinical trials for both human and other animals. We will also look at the epidemiology and control of vector insects. The journal will also cover the use of geographic information systems (Epi-GIS) for epidemiological surveillance which is a rapidly growing area of research in infectious diseases. Molecular epidemiological approaches are also particularly encouraged.