Machine Learning-based Classifiers for the Prediction of Low Birth Weight
1
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Objectives: Low birth weight (LBW) is a global concern associated with fetal and neonatal mortality as well as adverse consequencessuch as intellectual disability, impaired cognitive development, and chronic diseases in adulthood. Numerous factorscontribute to LBW and vary based on the region. The main objectives of this study were to compare four machine learningclassifiers in the prediction of LBW and to determine the most important factors related to this phenomenon in Hamadan,Iran. Methods: We carried out a retrospective cross-sectional study on a dataset collected from Fatemieh Hospital in 2017that included 741 mother-newborn pairs and 13 potential factors. Decision tree, random forest, artificial neural network,support vector machine, and logistic regression (LR) methods were used to predict LBW, with five evaluation criteria utilizedto compare performance. Results: Our findings revealed a 7% prevalence of LBW. The average accuracy of all models was87% or higher. The LR method provided a sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and accuracyof 74%, 89%, 7.04%, 29%, and 88%, respectively. Using LR, gestational age, number of abortions, gravida, consanguinity,maternal age at delivery, and neonatal sex were determined to be the six most important variables associated with LBW. Conclusions: Our findings underscore the importance of facilitating timely diagnosis of causes of abortion, providing geneticcounseling to consanguineous couples, and strengthening care before and during pregnancy (particularly for young mothers)to reduce LBW.
Abstract
Objectives: Low birth weight (LBW) is a global concern associated with fetal and neonatal mortality as well as adverse consequencessuch as intellectual disability, impaired cognitive development, and chronic diseases in adulthood. Numerous factorscontribute to LBW and vary based on the region. The main objectives of this study were to compare four machine learningclassifiers in the prediction of LBW and to determine the most important factors related to this phenomenon in Hamadan,Iran. Methods: We carried out a retrospective cross-sectional study on a dataset collected from Fatemieh Hospital in 2017that included 741 mother-newborn pairs and 13 potential factors. Decision tree, random forest, artificial neural network,support vector machine, and logistic regression (LR) methods were used to predict LBW, with five evaluation criteria utilizedto compare performance. Results: Our findings revealed a 7% prevalence of LBW. The average accuracy of all models was87% or higher. The LR method provided a sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and accuracyof 74%, 89%, 7.04%, 29%, and 88%, respectively. Using LR, gestational age, number of abortions, gravida, consanguinity,maternal age at delivery, and neonatal sex were determined to be the six most important variables associated with LBW. Conclusions: Our findings underscore the importance of facilitating timely diagnosis of causes of abortion, providing geneticcounseling to consanguineous couples, and strengthening care before and during pregnancy (particularly for young mothers)to reduce LBW.
참고문헌 (30)
- Talie A, "Magnitude of low birth weight and associated factors among newborns delivered in Dangla Primary Hospital, Amhara Regional State, Northwest Ethiopia, 2017", J Pregnancy, 2019, pp.3587239~ (2019)[DOI]
- "Global nutrition targets 2025: low birth weight policy brief" (2014)[URL]
- Ranjbaran M, "Prevalence of low birth weight and some associated factors in Markazi province, 2013-2014", World J Med Sci, 12(3), pp.252~258 (2015)[DOI]
- Sabbaghchi M, "A systematic review and meta-analysis on the prevalence of low birth weight infants in Iran", J Pregnancy, 2020, pp.3686471~ (2020)[DOI]
- "Low birth weight" (2023)[URL]
- Demelash H, "Risk factors for low birth weight in Bale zone hospitals, South-East Ethiopia : a case-control study", BMC Pregnancy Childbirth, 15, pp.264~ (2015)[DOI]
- "UNICEF-WHO low birthweight estimates: levels and trends 2000-2015" (2019)[URL]
- Ahmadi-Jouybari T, "Investigating factors affecting the interval between a burn and the start of treatment using data mining methods and logistic regression", BMC Med Res Methodol, 21(1), pp.71~ (2021)[DOI]
- Najafi-Ghobadi S, "Application of data mining techniques and logistic regression to model drug use transition to injection : a case study in drug use treatment centers in Kermanshah Province, Iran", Subst Abuse Treat Prev Policy, 14(1), pp.55~ (2019)[DOI]
- Eliyati N, "Support vector machines for classification of low birth weight in Indonesia", J Phys Conf Ser, 1282, pp.012010~ (2019)[DOI]
- Chawla NV, "SMOTE : synthetic minority over-sampling technique", J Artif Intell Res, 16, pp.321~357 (2002)[DOI]
- Dreiseitl S, "Logistic regression and artificial neural network classification models : a methodology review", J Biomed Inform, 35(5-6), pp.352~359 (2002)[DOI]
- abdulaziz Mohsen A, "New approach to medical diagnosis using artificial neural network and decision tree algorithm : application to dental diseases", Int J Inf Eng Electron Bus, 11(4), pp.52~60 (2019)[DOI]
- Sandhu AK, "Software reuse analytics using integrated random forest and gradient boosting machine learning algorithm", Softw Pract Exp, 51(4), pp.735~747 (2021)[DOI]
- Suzuki K, "Artificial neural networks: methodological advances and biomedical applications" (2011), InTech
- Huang S, "Applications of support vector machine(SVM)learning in cancer genomics", Cancer Genomics Proteomics, 15(1), pp.41~51 (2018)[DOI]
- Alanazi HO, "A critical review for developing accurate and dynamic predictive models using machine learning methods in medicine and health care", J Med Syst, 41(4), pp.69~ (2017)[DOI]
- Therneau TM, "An introduction to recursive partitioning using the RPART routines" (1997), Mayo Foundation
- Gevrey M, "Review and comparison of methods to study the contribution of variables in artificial neural network models", Ecol Model, 160(3), pp.249~264 (2003)[DOI]
- Khan W, "Infant birth weight estimation and low birth weight classification in United Arab Emirates using machine learning algorithms", Sci Rep, 12(1), pp.12110~ (2022)[DOI]
- Zahirzada A, "Implementing predictive model for low birth weight in Afghanistan", Proceedings of 2021 13th International Conference on Knowledge and Smart Technology (KST), pp.67~72 (2021)
- Borson NS, "Correlation analysis of demographic factors on low birth weight and prediction modeling using machine learning techniques", Proceedings of 2020 4th World Conference on Smart Trends in Systems, Security and Sustainability (WorldS4), pp.169~173 (2020)
- Senthilkumar D, "Prediction of low birth weight infants and its risk factors using data mining techniques", Proceedings of the 2015 International Conference on Industrial Engineering and Operations Management, pp.186~194 (2015)
- Brown JS, "Previous abortion and the risk of low birth weight and preterm births", J Epidemiol Community Health, 62(1), pp.16~22 (2008)[DOI]
- Ghelichkhani S, "Evaluation of maternal risk factors for preterm delivery in Fatemieh Hospital of Hamadan, Iran, 2019 : a case-control study", J Family Med Prim Care, 10(10), pp.3832~3837 (2021)[DOI]
- Cogendez E, "Post-abortion hysteroscopy : a method for early diagnosis of congenital and acquired intrauterine causes of abortions", Eur J Obstet Gynecol Reprod Biol, 156(1), pp.101~104 (2011)[DOI]
- Poorolajal J, "Effect of consanguinity on low birth weight : a meta-analysis", Arch Iran Med, 20(3), pp.178~184 (2017)
- Bennett RL, "Genetic counseling and screening of consanguineous couples and their offspring : recommendations of the National Society of Genetic Counselors", J Genet Couns, 11(2), pp.97~119 (2002)[DOI]
- Friede A, "Young maternal age and infant mortality : the role of low birth weight", Public Health Rep, 102(2), pp.192~199 (1987)
- Kebede A, "Trends and determinants of small birth weight in Ethiopia : further analysis of Ethiopian Demographic and Health Surveys", Ethiop J Health Sci, 31(2), pp.299~310 (2021)[DOI]