Prediction of Preeclampsia and Intrauterine Growth Restriction: Development of Machine Learning Models on a Prospective Cohort

Abstract

Background Preeclampsia and intrauterine growth restriction are placental dysfunction–related disorders (PDDs) that require a referral decision be made within a certain time period. An appropriate prediction model should be developed for these diseases. However, previous models did not demonstrate robust performances and/or they were developed from datasets with highly imbalanced classes. Objective In this study, we developed a predictive model of PDDs by machine learning that uses features at 24-37 weeks’ gestation, including maternal characteristics, uterine artery (UtA) Doppler measures, soluble fms-like tyrosine kinase receptor-1 (sFlt-1), and placental growth factor (PlGF). Methods A public dataset was taken from a prospective cohort study that included pregnant women with PDDs (66/95, 69%) and a control group (29/95, 31%). Preliminary selection of features was based on a statistical analysis using SAS 9.4 (SAS Institute). We used Weka (Waikato Environment for Knowledge Analysis) 3.8.3 (The University of Waikato, Hamilton, NZ) to automatically select the best model using its optimization algorithm. We also manually selected the best of 23 white-box models. Models, including those from recent studies, were also compared by interval estimation of evaluation metrics. We used the Matthew correlation coefficient (MCC) as the main metric. It is not overoptimistic to evaluate the performance of a prediction model developed from a dataset with a class imbalance. Repeated 10-fold cross-validation was applied. Results The classification via regression model was chosen as the best model. Our model had a robust MCC (.93, 95% CI .87-1.00, vs .64, 95% CI .57-.71) and specificity (100%, 95% CI 100-100, vs 90%, 95% CI 90-90) compared to each metric of the best models from recent studies. The sensitivity of this model was not inferior (95%, 95% CI 91-100, vs 100%, 95% CI 92-100). The area under the receiver operating characteristic curve was also competitive (0.970, 95% CI 0.966-0.974, vs 0.987, 95% CI 0.980-0.994). Features in the best model were maternal weight, BMI, pulsatility index of the UtA, sFlt-1, and PlGF. The most important feature was the sFlt-1/PlGF ratio. This model used an M5P algorithm consisting of a decision tree and four linear models with different thresholds. Our study was also better than the best ones among recent studies in terms of the class balance and the size of the case class (66/95, 69%, vs 27/239, 11.3%). Conclusions Our model had a robust predictive performance. It was also developed to deal with the problem of a class imbalance. In the context of clinical management, this model may improve maternal mortality and neonatal morbidity and reduce health care costs.

Data-Driven Modeling of Pregnancy-Related Complications

“…Machine learning is well suited for predictive modeling of pregnancy outcomes [91,92] and is becoming more prevalent [93][94][95][96][97][98][99][100][101][102][103][104][105] due to its ability to model highly complex relationships between measured features and outcomes. The majority of previous work focused on modeling techniques that incorporate one or two data sources, including clinical [95] as well as derived numerical data from another source such as blood samples [39], Doppler ultrasound, echosonography, or magnetic resonance imaging (MRI) readings [101], or mental health assessments [106]. Together, these datasets are combined into a single, structured table of samples and features.…”

Pre-Pregnancy Obesity vs. Other Risk Factors in Probability Models of Preeclampsia and Gestational Hypertension

“…Early PE (onset <34th week) is characterized by a more severe course, more frequent complications in the mother, and complications in the fetus (intrauterine growth restriction (IUGR), as well as fetal hypoxia or death of the fetus), which may suggest a significant contribution of placental circulation pathology. In late PE, the placenta is often normal; therefore, this form of the disease may be characterized by a lower frequency of fetal complications [13,16,[27][28][29]. It was also found that PE manifesting mainly maternal symptoms may display risk factors similar to those of GH (obesity, insulin-resistance, hyperlipidemia, and chronic hypertension), which may suggest the presence of disorders in the mother's blood vessels before pregnancy [13].…”

“…The etiology of pregnancy-induced hypertension is not fully explained, but the pathogenesis of PE takes into account disturbances in trophoblast invasion into the walls of the spiral arteries, which results in a lack of remodeling, high-resistance circulation, and placental ischemia [37,38]. The balance of many placental biomarkers is disturbed, oxidative stress and inflammation in both the placenta and in the mother's circulation are intensified, and the effect is endothelial function damage and an increase in blood pressure [16,[37][38][39]. There are many studies available in the literature which have found associations between lower levels of antioxidants in pregnant women with PE [40].…”

“…Pregnancy-induced hypertension occurs, on average, in 10% of the pregnant population, but, in some regions of the world, its incidence is higher [ 14 , 15 ]. Hypertension in pregnancy, especially preeclampsia (PE), is one of the main causes of mortality and morbidity in mothers and their babies, and the effects of the disease are not only limited to the perinatal period, but also have long-term effects related to, e.g., a future higher risk of cardiovascular disease in both mothers and babies [ 13 , 15 , 16 , 17 , 18 , 19 ]. Early identification of women at risk of PIH is needed so they can be placed under increased surveillance as early as possible.…”

Prediction of hypertensive disorders in pregnancy based on placental growth factor

“…With the development of various aspects of technology and engineering, in addition to statistical prediction methods, there are also artificial intelligence algorithms such as machine learning. Sufriyana et al used maternal characteristics, uterine artery (UtA) doppler measurement, sFlt-1 and PlGF in the second and third trimester of pregnancy to study the machine learning related model for predicting preeclampsia (PE) and intrauterine growth restriction [ 9 ]. Vascular biomarkers have been conducted many different types of comparative analysis [ 10 ].…”