Comparison of artificial neural network and logistic regression model for factors affecting birth weight

• Published in: SN applied sciences Vol. 1; no. 4; p. 378
• Main Author: Kirişci, Murat
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01-04-2019; Springer Nature B.V
• Subjects: Applied and Technical Physics; Artificial neural networks; Babies; Birth weight; Body mass index; Caffeine; Chemistry/Food Science; Computers; Data acquisition; Decision making; Earth Sciences; Engineering; Engineering: Artificial Intelligence; Environment; Fuzzy sets; Independent variables; Information storage; Materials Science; Morbidity; Mortality; Neighborhoods; Neural networks; Pregnancy; Prenatal care; Public health; Regression analysis; Regression models; Research Article; Risk factors; Rural areas; Istanbul Turkey; Turkey; Logistic regression; Birth weight; Neonate; Artificial neural network; ROC curve
• ISSN: 2523-3963; 2523-3971
• DOI: 10.1007/s42452-019-0391-x

The aim of this work compares the ANN and logistic regression analysis to determine the factors affecting birth weight. This study included 223 newborn babies. The records of babies born between January 2017 and December 2017 were used. The data were obtained from Beykoz district of Istanbul. ANN and logistic regression analysis of the method obtained based on these records were evaluated. Logistic regression revealed the items GB, MA, GA, NH, BMI, MPPW, MWGP, MsAU, MsCU, MsE as significant factors for BW. The area under the receiver operating characteristic (AuROC) curve 0.941 (SD = 0.0012) for ANN and 0.909 (SD = 0.019) for Logistic Regression model. The ANNs may be trained with data acquired in various contexts and can consider local expertise, differences, and other variables with uncertain effects on outcome. Although the ANN value is greater than the LR value, these results are very close to each other. This shows us that in terms of their classification ability, these two methods are approximately equal to each other. The results we have seen in our study show that in the medical diagnosis, neither model can change the other. Both models can be used as a complement to help with decision-making. Both models have the potential to help physicians with respect to understanding BW risk factors, risk estimation.