Comparison of logistic regression with machine learning methods for the prediction of fetal growth abnormalities: a retrospective cohort study

Abstract

BackgroundWhile there is increasing interest in identifying pregnancies at risk for adverse outcome, existing prediction models have not adequately assessed population-based risks, and have been based on conventional regression methods. The objective of the current study was to identify predictors of fetal growth abnormalities using logistic regression and machine learning methods, and compare diagnostic properties in a population-based sample of infants.MethodsData for 30,705 singleton infants born between 2009 and 2014 to mothers resident in Nova Scotia, Canada was obtained from the Nova Scotia Atlee Perinatal Database. Primary outcomes were small (SGA) and large for gestational age (LGA). Maternal characteristics pre-pregnancy and at 26 weeks were studied as predictors. Logistic regression and select machine learning methods were used to build the models, stratified by parity. Area under the curve was used to compare the models; relative importance of predictors was compared qualitatively.Results7.9% and 13.5% of infants were SGA and LGA, respectively; 48.6% of births were to primiparous women and 51.4% were to multiparous women. Prediction of SGA and LGA was poor to fair (area under the curve 60–75%) and improved with increasing parity and pregnancy information. Smoking, previous low birthweight infant, and gestational weight gain were important predictors for SGA; pre-pregnancy body mass index, gestational weight gain, and previous macrosomic infant were the strongest predictors for LGA.ConclusionsThe machine learning methods used in this study did not offer any advantage over logistic regression in the prediction of fetal growth abnormalities. Prediction accuracy for SGA and LGA based on maternal information is poor for primiparous women and fair for multiparous women.Electronic supplementary materialThe online version of this article (10.1186/s12884-018-1971-2) contains supplementary material, which is available to authorized users.

Functional Outcome Prediction in Ischemic Stroke: A Comparison of Machine Learning Algorithms and Regression Models

“…Our analyses revealed that ML algorithms and logistic regression models had comparable predictive accuracy when validated internally and externally. Our findings buttress current evidence from other published studies (28)(29)(30)(31)(32)(33) that already showed that the logistic regression and ML algorithms had comparable predictive accuracy in empirical clinical studies. A recently published systematic review found no evidence of the superior predictive performance of ML models over logistic regression models in clinical studies (32).…”

Prediction of perinatal death using machine learning models: a birth registry-based cohort study in northern Tanzania

“…7 Early identification of pregnant women at risk for adverse maternal and perinatal outcomes during the prenatal period and timely provision of high-quality healthcare services have been reported to improve maternal and newborn survival. 9 Machine learning (hereafter denoted as 'ML') models are methodologies for developing algorithms that learn from existing data to make predictions on new data. 9 ML models have shown better predictive performance over the classical or conventional regression models, 10 and they can better handle a significant number of potential predictors.…”

“…9 Machine learning (hereafter denoted as 'ML') models are methodologies for developing algorithms that learn from existing data to make predictions on new data. 9 ML models have shown better predictive performance over the classical or conventional regression models, 10 and they can better handle a significant number of potential predictors. However, there is conflicting evidence of the performance of these models.…”

Machine Learning in Fetal Cardiology: What to Expect

“…Similarly, Rawat et al [64] implemented an ANN model using again 2-D US morphometric measurements from a total of 120 fetuses. Recently, Kuhle et al [65] compared different ML methods to predict fetal growth abnormalities in a cohort of more than 30,000 patients. However, the authors reported that the ML methods used did not offer any advantage over logistic regression in the prediction of fetal growth abnormalities.…”