Application of machine learning methods for predicting infant mortality in Rwanda: analysis of Rwanda demographic health survey 2014–15 dataset

Abstract

Background Extensive research on infant mortality (IM) exists in developing countries; however, most of the methods applied thus far relied on conventional regression analyses with limited prediction capability. Advanced of Machine Learning (AML) methods provide accurate prediction of IM; however, there is no study conducted using ML methods in Rwanda. This study, therefore, applied Machine Learning Methods for predicting infant mortality in Rwanda. Methods A cross-sectional study design was conducted using the 2014–15 Rwanda Demographic and Health Survey. Python software version 3.8 was employed to test and apply ML methods through Random Forest (RF), Decision Tree, Support Vector Machine and Logistic regression. STATA version 13 was used for analysing conventional methods. Evaluation metrics methods specifically confusion matrix, accuracy, precision, recall, F1 score, and Area under the Receiver Operating Characteristics (AUROC) were used to evaluate the performance of predictive models. Results Ability of prediction was between 68.6% and 61.5% for AML. We preferred with the RF model (61.5%) presenting the best performance. The RF model was the best predictive model of IM with accuracy (84.3%), recall (91.3%), precision (80.3%), F1 score (85.5%), and AUROC (84.2%); followed by decision tree model with model accuracy (83%), recall (91%), precision (79%), F1 score (84.67%) and AUROC(82.9%), followed by support vector machine with model accuracy (68.6%), recall (74.9%), precision(67%), F1 score (70.73%) and AUROC (68.6%) and last was a logistic regression with the low accuracy of prediction (61.5%), recall (61.1%), precision (62.2%), F1 score (61.6%) and AUROC (61.5%) compared to other predictive models. Our predictive models showed that marital status, children ever born, birth order and wealth index are the 4 top predictors of IM. Conclusions In developing a predictive model, ML methods are used to classify certain hidden information that could not be detected by traditional statistical methods. Random Forest was classified as the best classifier to be used for the predictive models of IM.

Community and individual level determinants of infant mortality in rural Ethiopia using data from 2016 Ethiopian demographic and health survey

“…Separated women had a higher risk of infant mortality than married women. This finding is in agreement with a study from Rwanda 40 and United States 41 .This could be due to socioeconomic issues, cultural norms, and the lifestyle consequences of single women. Similarly, infants born with a small birth size had a higher risk of infant mortality than infants born with a large birth size.…”

Predicting perinatal mortality based on maternal health status and health insurance service using homogeneous ensemble machine learning methods

“…The explanatory analysis shows that perinatal mortality has been reducing over time in Ethiopia due to an increase in the number of hospitals in rural areas and the introduction of community-based health insurance which encourages pregnant women to visit the hospital to give birth. The proposed model, which was developed using a gradient boosting machine learning algorithm, achieved an overall performance of 90.24%, which is a better result compared to a result achieved by previous studies [17,20,24] which achieved 83% overall performance using gradient boosting machine learning algorithm. Based on the feature importance analysis, perinatal mortality in Ethiopia is significantly associated with health insurance and maternal health status such as birth interval of less than 2 years, preterm delivery, anemia, congenital anomaly, educational status, family size, occupation, marital status, perceived quality of care, the first choice of place for treatment during illness, previous history of perinatal death, not receiving tetanus toxoid immunization, and lack of iron supplementation.…”

“…This study, hence, is motivated to fill these gaps by constructing a predictive model, identifying risk factors, designing artifacts, and generating relevant rules that help to develop evidence-based strategies, policies and interventions towards preventing, controlling and/or ending perinatal mortality in Ethiopia. Emmanuel et al [ 20 ] explored application of machine learning methods for predicting infant mortality in Rwanda using Rwanda demographic health survey 2014–15 dataset and showed effectiveness and explain ability of machine learning algorithms such as Random forest, decision tree, support vector machine and logistic regression. For this reason, ensemble machine learning algorithms were selected for experiment in this study.…”

Employing supervised machine learning algorithms for classification and prediction of anemia among youth girls in Ethiopia

“…Machine learning methods possess the capacity to surpass traditional statistical approaches by effectively managing extensive and intricate nonlinear data, operating without the need for preexisting assumptions, and capturing intricate connections among predictors 17 , 18 . Overall, the utilization of machine learning algorithms for classification and prediction offers numerous advantages, including automation, pattern recognition, adaptability, scalability, objectivity, handling non-linearity, feature selection, and generalization.…”