Using Machine Learning to Predict Complications in Pregnancy: A Systematic Review

Abstract

Introduction: Artificial intelligence is widely used in medical field, and machine learning has been increasingly used in health care, prediction, and diagnosis and as a method of determining priority. Machine learning methods have been features of several tools in the fields of obstetrics and childcare. This present review aims to summarize the machine learning techniques to predict perinatal complications.Objective: To identify the applicability and performance of machine learning methods used to identify pregnancy complications.Methods: A total of 98 articles were obtained with the keywords “machine learning,” “deep learning,” “artificial intelligence,” and accordingly as they related to perinatal complications (“complications in pregnancy,” “pregnancy complications”) from three scientific databases: PubMed, Scopus, and Web of Science. These were managed on the Mendeley platform and classified using the PRISMA method.Results: A total of 31 articles were selected after elimination according to inclusion and exclusion criteria. The features used to predict perinatal complications were primarily electronic medical records (48%), medical images (29%), and biological markers (19%), while 4% were based on other types of features, such as sensors and fetal heart rate. The main perinatal complications considered in the application of machine learning thus far are pre-eclampsia and prematurity. In the 31 studies, a total of sixteen complications were predicted. The main precision metric used is the AUC. The machine learning methods with the best results were the prediction of prematurity from medical images using the support vector machine technique, with an accuracy of 95.7%, and the prediction of neonatal mortality with the XGBoost technique, with 99.7% accuracy.Conclusion: It is important to continue promoting this area of research and promote solutions with multicenter clinical applicability through machine learning to reduce perinatal complications. This systematic review contributes significantly to the specialized literature on artificial intelligence and women’s health.

Novel Early Pregnancy Multimarker Screening Test for Preeclampsia Risk Prediction

“…learning approach to select statistically most significant PE prediction models (22)(23)(24), adjusting biomarker measurements for the gestational day at sampling, and incorporation of the placental genotypes of the FLT1 rs4769613 T/C variant, confidently associated with PE susceptibility (17,18). The prediction model combining gestational age-adjusted 6PLEX measurements of PTX3, sFlt-1, sENG, and ADAM12 with parity and placental FLT1 rs4769613 T/C genotype data yielded a correct prediction of PE in 93.5% of analyzed cases with no false-negative predictions.…”

The use of artificial intelligence and machine learning methods in first trimester pre-eclampsia screening: a systematic review protocol

“…Studies have been performed which use AI/ML methods to predict complications in pregnancy [24]. However, as most studies focus on second trimester pregnancies an overview of first trimester pregnancy PE prediction using AI/ML methods is needed as identification of PE risk in the first trimester can be used to select pregnancies that may benefit from preventive treatment with Low dose aspirin, which has been documented to reduce the occurrence of PE [7, 30].…”

“…By enabling the combination of clinical phenotype information (extracted from electronic health records) and biomarker information along with environmental exposures, AI/ML methods provide the promise of producing a clinically relevant PE prediction algorithm [20]. Furthermore, AI/ML methods have been applied to the prediction of pregnancy complications, albeit not in a clinical care setting [24]. The objective of this systematic review is to identify and assess studies regarding the application of AI/ML methods in first-trimester screening for PE.…”

Symbolic regression analysis of interactions between first trimester maternal serum adipokines in pregnancies which develop pre-eclampsia

“…Frequently based on assumptions about the distribution of markers as a function of gestational age and basic clinical information [14, 16, 19]. Machine learning (ML) models, with their promise of integrating multiple “omics” layers [29] and, potential to revolutionise individual patient treatment [30], have recently been introduced, albeit not yet in diagnostic use [31]. ML procedures are frequently of a “black-box” type, not revealing any underlying organization of parameters allowing for interpretability [32].…”