Predicting risks of low birth weight in Bangladesh with machine learning

Pollob, S.M. Ashikul Islam; Abedin, Md. Menhazul; Islam, Md. Touhidul; Islam, Md. Merajul; Maniruzzaman, Md.

doi:10.1371/journal.pone.0267190

Cited by 31 publications

(23 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Covariates included in our multivariable models were chosen a priori based on relevant literature (Islam Pollob et al 2022 ; Khan et al 2018 ; Kiserud et al 2017 ; Shah et al 2014 ; Tamirat et al 2021 ; Zhou et al 2016 ). Our models included the following potential confounders: maternal age (years), height (cm), weight (kg), education level (six ordered categories), parity ( n ), history of pregnancy loss (yes; no), antenatal attendance (yes; no), iron supplementation (yes; no), infant sex (male, female), household wealth index (quintiles, with first two groups collapsed), average dietary diversity score ( 10 food group scale ), and food security status (food secure; food insecure).…”

Section: Methodsmentioning

confidence: 99%

Maternal exposure to multiple mycotoxins and adverse pregnancy outcomes: a prospective cohort study in rural Bangladesh

et al. 2023

View full text Add to dashboard Cite

There is limited and inconsistent evidence, primarily from cross-sectional studies, linking mycotoxins to adverse birth outcomes. This study investigates the potential role of maternal dietary exposure to multiple mycotoxins in the development of several adverse pregnancy and birth outcomes. We analyzed data from 436 singleton pregnancies enrolled in a prospective cohort study in the rural Habiganj district, Bangladesh, between July 2018 and November 2019. Thirty-five urinary mycotoxin biomarkers were quantified using liquid chromatography coupled with tandem mass spectrometry and used to estimate dietary mycotoxin exposure. Multivariable regression models, adjusted for potential confounding and clustering, were fitted to assess the associations between maternal exposure to frequently occurring mycotoxins (ochratoxin A-OTA, citrinin- CIT, and Deoxynivalenol- DON) and pregnancy loss, preterm birth (PTB), low birth weight (LBW), born small-for-gestational-age (SGA) and small-vulnerable newborn. The results indicate that only in 16 of 436 pregnancies (4%) were urine samples free from all investigated mycotoxins. Biomarkers for six major mycotoxins were detected in the urine samples. OTA (95%), CIT (61%), and DON (6%) were most frequently detected, with at least two mycotoxins co-occurring in the majority of women (63%). There was evidence that maternal dietary intake of OTA was associated with higher odds of having an LBW baby, with the odds increasing in a dose-dependent manner. We found no evidence of associations between pregnancy loss, PTB, SGA, small-vulnerable newborns, and maternal dietary exposure to OTA, CIT, and DON, albeit with large confidence intervals, so findings are consistent with protective as well as large harmful effects. Exposure to multiple mycotoxins during pregnancy is widespread in this rural community and represents a health risk for mothers and babies. Tailored public health policies and interventions must be implemented to reduce mycotoxin exposure to the lowest possible level.

show abstract

Section: Methodsmentioning

confidence: 99%

Maternal exposure to multiple mycotoxins and adverse pregnancy outcomes: a prospective cohort study in rural Bangladesh

et al. 2023

View full text Add to dashboard Cite

show abstract

“…Birth weight was classified as a binary variable by using a threshold of 2.5 kg at the time of the baby’s birth. If a baby’s birth weight was less than 2.5 kilograms, it is then classified as low birth weight (LBW) [ 69 ]. One additional dummy variable was created to represent the caregivers’ knowledge of dietary diversity, which refers to their understanding of proper nutrition while caring for the child in the absence of the mother.…”

Section: Methodsmentioning

confidence: 99%

Nutritional status of under-five aged children of ready-made garment workers in Bangladesh: A cross-sectional study

Haque

Rafi²,

Zaman³

et al. 2023

PLoS ONE

View full text Add to dashboard Cite

Background The ready-made garment (RMG) sector is a significant contributor to the economic growth of Bangladesh, accounting for 10% of the country’s GDP and more than 80% of its foreign exchange earnings. The workforce in this sector is predominantly made up of women, with 2.5 million women working in the industry. However, these women face numerous challenges in carrying out their culturally-expected household responsibilities, including childcare, due to severe resource constraints. As a result, the children of these working women have a higher incidence of malnutrition, particularly stunted growth. This study aims to identify the factors that contribute to stunting in children under the age of five whose mothers work in the RMG sector in Bangladesh. Methods The study collected data from 267 female RMG workers in the Gazipur district of Bangladesh using a simple random sampling technique. Chi-square tests were used to determine the associations between the factors influencing child stunting, and Multinomial Logit Models were used to estimate the prevalence of these factors. Results The study found that the prevalence of moderate and severe stunting among the children of RMG workers living in the Gazipur RMG hub was 19% and 20%, respectively. The study identified several significant predictors of child stunting, including the mother’s education level, nutritional knowledge, control over resources, receipt of antenatal care, household size, sanitation facilities, and childbirth weight. The study found that improving the mother’s education level, increasing household size, and receiving antenatal care during pregnancy were important factors in reducing the likelihood of child stunting. For example, if a mother’s education level increased from no education to primary or secondary level, the child would be 0.211 (0.071–0.627) and 0.384 (0.138–1.065) times more likely to have a normal weight and height, respectively, than to be moderately stunted. Conclusion The study highlights the challenges faced by working women in the RMG sector, who often receive minimal wages and have limited access to antenatal care services. To address these challenges, the study recommends policies that support antenatal care for working-class mothers, provide daycare facilities for their children, and implement a comprehensive social safety net program that targets child nutrition. Improving the socioeconomic status of mothers is also critical to reducing child malnutrition in this population.

show abstract

“…Several studies have been conducted in Asia to predict low birth weight in infants. In Bangladesh, a study [3] showed that logistic regression (LR) is more effective than decision tree (DT) in predicting low birth weight (LBW), with an accuracy of 85% using a 70:30 training and test dataset ratio. In India, researchers claim that the Classi cation Tree algorithm had the highest overall prediction accuracy, speci city, AUC, Fvalue, and Precision compared to other classi cation methods, with an accuracy rate of 89.95% [17].…”

Section: Prediction Of Lbwmentioning

confidence: 99%

“…LBW is a major healthcare concern in Bangladesh, with a prevalence rate of 17.7% in 2011, 20% in 2014, and 16% in 2017. Researchers used the chi-square test, DT, and LR models to predict LBW [3]. Evidently, the prevalence of LBW in Bangladesh consistently exceeds 10%, indicating the utmost importance of eradicating this issue.…”

Section: Introductionmentioning

confidence: 99%

Prediction and Feature selection of Low Birth Weight using Machine Learning Algorithms

Reza,

Salma

2024

Preprint

View full text Add to dashboard Cite

Background and Aims: The birth weight of a newborn is a crucial factor that affects their overall health and future well-being. Low birth weight (LBW) is a widespread global issue, which the World Health Organization defines as weighing less than 2,500 grams. LBW can have severe negative consequences on an individual's health, including neonatal mortality and various health concerns throughout their life. To address this problem, this study has been conducted using BDHS 2017-2018 data to identify the most relevant features of LBW and the best model. Methods: The study used data extracted from BDHS 2017-2018. The Boruta algorithm and Wrapper method were employed to identify essential features. Several machine learning classifiers were then used, including Decision Tree, SVM (Support Vector Machine), Naïve Bayes, Random Forest, XGBoost (eXtreme Gradient Boosting), and AdaBoost (Adaptive Boosting), to determine the best model for predicting LBW. Results: According to the study, the Boruta algorithm identifies eleven significant features, which include the respondent's age, highest education level, educational attainment, wealth index, age at first birth, weight, height, BMI, age at first sexual intercourse, birth order number, and whether the child is a twin. The machine-learning classifiers used in the study were DT, SVM, NB, RF, XGBoost, and AB models, which had AUCs of 0.538, 0.532, 0.529, 0.549, 0.507, and 0.509, respectively. The DT and RF models, implemented using the wrapper method, identified three significant features: the region, whether the child is a twin, and delivery by cesarean section, with AUC values of 0.5550 and 0.5749, respectively. On the other hand, the SVM, NB, and AB models identified "child is twin" as a significant feature with an AUC value of 0.5120. Finally, the XGBoost model identified "age at 1st sex" and "child is twin" as significant features with an AUC value of 0.508. Conclusions: Based on the analysis, the authors suggest that Random Forest (RF) is the most effective predictive model for Low Birth Weight (LBW). They also found that the Wrapper method is the best feature selection technique. Notable features of LBW include child is a twin, region, and delivery by caesarean section.

show abstract

Predicting risks of low birth weight in Bangladesh with machine learning

Cited by 31 publications

References 37 publications

Maternal exposure to multiple mycotoxins and adverse pregnancy outcomes: a prospective cohort study in rural Bangladesh

Maternal exposure to multiple mycotoxins and adverse pregnancy outcomes: a prospective cohort study in rural Bangladesh

Nutritional status of under-five aged children of ready-made garment workers in Bangladesh: A cross-sectional study

Prediction and Feature selection of Low Birth Weight using Machine Learning Algorithms

Contact Info

Product

Resources

About