Using best performance machine learning algorithm to predict child death before celebrating their fifth birthday

Workie Demsash, Addisalem

doi:10.1016/j.imu.2023.101298

Cited by 5 publications

(6 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The infants born from rich mothers had a 40% lower risk of death compared with those whose mothers were poor, respectively. This evidence agreed with studies done in Ethiopia [ 67 , 68 ], Bangladesh [ 69 , 70 ], Pakistan [ 71 ], and Nigeria [ 72 ]. This might be because children of the richest mothers have a lower chance of getting anemia than those of poor mothers; richest mothers are more likely to provide balanced and diversified food to their infants [ 73 ]; and they might have a better chance of accessing healthcare services than economically poor mothers [ 74 , 75 ].…”

Section: Discussionsupporting

confidence: 92%

Mothers’ experience of losing infants by death and its predictors in Ethiopia

Demsash,

Asefa,

Bekana

2024

PLoS ONE

View full text Add to dashboard Cite

Background Although infant deaths worldwide have reduced, many children die before their first birthday. Infant deaths are widespread in low-income countries, and information about the cause of death is limited. In Ethiopia, 53% of infants’ deaths occurred in their neonatal period, and 174 infants’ deaths occurred from 3684 births. Hence, this study aimed to assess mothers’ experiences with infant death and its predictors in Ethiopia. Methods A total of 1730 weighted samples of mothers from the 2019 EDHS dataset, which was collected across the regions of Ethiopia, were included for analysis. A two-stage cluster sampling technique with a cross-sectional study design was used. All mothers whose children were under the age of 0–12 months were included in this study. Six count regression models were considered and compared using Akaike’s information criteria and Bayesian information criterion with STATA version 15 software. The strength of the association between the number of infant deaths and possible predictors was determined at a P-value less than 0.05, with a 95% confidence interval. The findings were interpreted by using the incident rate ratio. Results A total of 46.3% of mothers had lost at least one infant by death in the last five years before the 2019 EDHS survey was held. The mean and variance of infant deaths were 2.55 and 5.58, respectively. The histogram was extremely picked at the beginning, indicating that a large number of mothers did not lose their infants by death, and that shows the data had positive skewness. Mothers under 25–29 years of age (IRR: 1.75, 95% CI:1.48, 2.24), and 30–34 years of age (IRR: 1.42, 95% CI: 1.12, 2.82), Somali (IRR: 1.47, 95% CI: 1.02, 3.57), Gambela (IRR: 1.33, 95% CI: 1.10, 2.61), and Harari (IRR: 1.39, 95% CI: 1.02, 2.63) regions, rural resident mothers (IRR: 1.68, 95% CI: 1.09, 1.91, and Protestant (IRR = 1.43, 95% CI: 1.14, 2.96), and Muslim (IRR = 1.59, 95% CI: 1.07, 2.62) religion fellow of mothers were associated with a high risk of infants’ deaths. Whereas, being rich IRR: 0.37, 95% CI: .27, .81) and adequate ANC visits (IRR: 0.28, 95% CI: .25, .83) were associated with a low risk of infant death. Conclusion Many mothers have experienced infant deaths, and the majority of infants’ deaths occur after the first month of birth. Encouraging mothers to attend antenatal care visits, creating mothers’ awareness about childcare, and ensuring equal health services distribution and utilization to rural residents are essential to minimize infant death. Educating lower-aged reproductive mothers would be a necessary intervention to prevent and control infant deaths.

show abstract

Section: Discussionsupporting

confidence: 92%

Mothers’ experience of losing infants by death and its predictors in Ethiopia

Demsash,

Asefa,

Bekana

2024

PLoS ONE

View full text Add to dashboard Cite

show abstract

“…The j48, multilayer perceptron, and random forest algorithms were the second, third, and fourth best machine learning algorithms to predict childhood vaccination with 89.24%, 87.20%, and 82.37% accuracy, respectively. This finding was supported by various studies conducted to predict under-five child mortality [ 29 , 32 , 44 , 79 ], contraceptive discontinuation [ 70 ], stunting, and malnutrition among children [ 80 – 82 ].…”

Section: Discussionsupporting

confidence: 54%

“…Generating rules for childhood vaccination was the third objective of the study. Previous studies have assessed the joint effect of independent predictors on the outcome of interest [ 44 , 70 , 78 ]. Consequently, seven association rules were generated to determine vaccination status among children aged 12–23 months in Ethiopia.…”

Section: Discussionmentioning

confidence: 99%

“…In the second stage, a best-performance machine learning algorithm with information gain values was used to find important features or attributes that have a major contribution to predicting childhood vaccination among children aged 12–23 months in Ethiopia. The highest information gain value of an independent attribute is the most important attribute to predict childhood vaccination [ 44 ]. Then the next important attributes were selected based on their order of highest information gain value.…”

Section: Operationalizationsmentioning

confidence: 99%

“…The knowledge flow can handle data either incrementally or in batches. The features of knowledge flow are initiative of the data flow layout, processing the data in batches or incrementally, processing multiple batches or streams in parallel, chain filtering together, and viewing and visualized model performance with a fold cross-validation [ 44 ]. The overall knowledge flow of model building for data processing, analyzing, and visualizing has been presented in Fig 1 .…”

Section: Model Buildingmentioning

confidence: 99%

See 2 more Smart Citations

Machine learning algorithms’ application to predict childhood vaccination among children aged 12–23 months in Ethiopia: Evidence 2016 Ethiopian Demographic and Health Survey dataset

Demsash,

Chereka,

Walle

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

Introduction Childhood vaccination is a cost-effective public health intervention to reduce child mortality and morbidity. But, vaccination coverage remains low, and previous similar studies have not focused on machine learning algorithms to predict childhood vaccination. Therefore, knowledge extraction, association rule formulation, and discovering insights from hidden patterns in vaccination data are limited. Therefore, this study aimed to predict childhood vaccination among children aged 12–23 months using the best machine learning algorithm. Methods A cross-sectional study design with a two-stage sampling technique was used. A total of 1617 samples of living children aged 12–23 months were used from the 2016 Ethiopian Demographic and Health Survey dataset. The data was pre-processed, and 70% and 30% of the observations were used for training, and evaluating the model, respectively. Eight machine learning algorithms were included for consideration of model building and comparison. All the included algorithms were evaluated using confusion matrix elements. The synthetic minority oversampling technique was used for imbalanced data management. Informational gain value was used to select important attributes to predict childhood vaccination. The If/ then logical association was used to generate rules based on relationships among attributes, and Weka version 3.8.6 software was used to perform all the prediction analyses. Results PART was the first best machine learning algorithm to predict childhood vaccination with 95.53% accuracy. J48, multilayer perceptron, and random forest models were the consecutively best machine learning algorithms to predict childhood vaccination with 89.24%, 87.20%, and 82.37% accuracy, respectively. ANC visits, institutional delivery, health facility visits, higher education, and being rich were the top five attributes to predict childhood vaccination. A total of seven rules were generated that could jointly determine the magnitude of childhood vaccination. Of these, if wealth status = 3 (Rich), adequate ANC visits = 1 (yes), and residency = 2 (Urban), then the probability of childhood vaccination would be 86.73%. Conclusions The PART, J48, multilayer perceptron, and random forest algorithms were important algorithms for predicting childhood vaccination. The findings would provide insight into childhood vaccination and serve as a framework for further studies. Strengthening mothers’ ANC visits, institutional delivery, improving maternal education, and creating income opportunities for mothers could be important interventions to enhance childhood vaccination.

show abstract