The prediction of preeclampsia: the way forward

Prediction of pregnancy-related complications in women undergoing assisted reproduction, using machine learning methods

OBJECTIVE

To use machine learning methods to develop prediction models of pregnancy complications in women who conceived with assisted reproductive techniques (ART).

DESIGN

A nation-wide register-based cohort study with prospectively collected data.

SETTING

Swedish national registers and nationwide quality IVF register.

PATIENT(S)

all nulliparous women who achieved birth within the first 3 ART treatment cycles between 2008 and 2016 in Sweden.

INTERVENTION(S)

Characteristics before the use of ART, such as demographics and medical history, were considered potential predictors in the development of before treatment prediction models. ART treatment details were further included in after treatment prediction models.

MAIN OUTCOME MEASURE(S)

Potential diagnoses of preeclampsia, placental complications (previa, accreta, and abruption), and postpartum hemorrhage were identified using the International Classification of Diseases recorded in the Swedish Medical Birth and Patient registers, respectively. Multiple prediction model algorithms were performed and compared for each outcome and treatment cycle, including logistic regression, decision tree model, naïve Bayes classification, support vector machine, random forest, and gradient boosting. The performance of each model was assessed with C statistic, and nested cross-validation was used to aid model selection and hyperparameter tuning.

RESULT(S)

A total of 14,732 women gave birth after the first (N = 7,302), second (N = 4,688), or third (N = 2,742) ART cycle, representing birth rates of 24.1%, 23.8%, and 22.0%. Overall prediction performance did not vary much across the different methods used. In the first cycle, the before treatment prediction performance was at best 66%, 66%, and 60% for preeclampsia, placental complications, and postpartum hemorrhage, respectively. Inclusion of after treatment characteristics conferred slight improvement (approximately 1%-5%), as did prediction in later cycles (approximately 1%-5%). The top influential and consistent predictors included age, region of residence, infertility diagnosis, and type of embryo transfer (fresh or frozen) in the later (2nd and 3rd) cycles. Body mass index was a top predictor of preeclampsia and was also influential for placental complications but not for postpartum hemorrhage.

CONCLUSION(S)

The combined use of demographics, medical history, and ART treatment information was not enough to confidently predict serious pregnancy complications in women who conceived with ART. Future studies are needed to assess if additional longitudinal follow-up during pregnancy can improve the prediction to allow clinical protocol development.

Machine learning techniques for prediction in pregnancy complicated by autoimmune rheumatic diseases: Applications and challenges

Autoimmune rheumatic diseases are chronic conditions affecting multiple systems and often occurring in young women of childbearing age. The diseases and the physiological characteristics of pregnancy significantly impact maternal-fetal health and pregnancy outcomes. Currently, the integration of big data with healthcare has led to the increasing popularity of using machine learning (ML) to mine clinical data for studying pregnancy complications. In this review, we introduce the basics of ML and the recent advances and trends of ML in different prediction applications for common pregnancy complications by autoimmune rheumatic diseases. Finally, the challenges and future for enhancing the accuracy, reliability, and clinical applicability of ML in prediction have been discussed. This review will provide insights into the utilization of ML in identifying and assisting clinical decision-making for pregnancy complications, while also establishing a foundation for exploring comprehensive management strategies for pregnancy and enhancing maternal and child health.

AI-based preeclampsia detection and prediction with electrocardiogram data

Introduction

More than 76,000 women die yearly from preeclampsia and hypertensive disorders of pregnancy. Early diagnosis and management of preeclampsia can improve outcomes for both mother and baby. In this study, we developed artificial intelligence models to detect and predict preeclampsia from electrocardiograms (ECGs) in point-of-care settings.

Methods

Ten-second 12-lead ECG data was obtained from two large health care settings: University of Tennessee Health Science Center (UTHSC) and Atrium Health Wake Forest Baptist (AHWFB). UTHSC data was split into 80% training and 20% holdout data. The model used a modified ResNet convolutional neural network, taking one-dimensional raw ECG signals comprising 12 channels as an input, to predict risk of preeclampsia. Sub-analyses were performed to assess the predictive accuracy for preeclampsia prediction within 30, 60, or 90 days before diagnosis.

Results

The UTHSC cohort included 904 ECGs from 759 females (78.8% African American) with a mean ± sd age of 27.3 ± 5.0 years. The AHWFB cohort included 817 ECGs from 141 females (45.4 African American) with a mean ± sd age of 27.4 ± 5.9 years. The cross-validated ECG-AI model yielded an AUC (95% CI) of 0.85 (0.77-0.93) on UTHSC holdout data, and an AUC (95% CI) of 0.81 (0.77-0.84) on AHWFB data. The sub-analysis of different time windows before preeclampsia prediction resulted in AUCs (95% CI) of 0.92 (0.84-1.00), 0.89 (0.81-0.98) and 0.90 (0.81-0.98) when tested on ECGs 30 days, 60 days and 90 days, respectively, before diagnosis. When assessed on early onset preeclampsia (preeclampsia diagnosed at <34 weeks of pregnancy), the model's AUC (95% CI) was 0.98 (0.89-1.00).

Discussion

We conclude that preeclampsia can be identified with high accuracy via application of AI models to ECG data.

Alterations in the von Willebrand factor/ADAMTS-13 axis in preeclampsia

BACKGROUND

Preeclampsia is a gestational hypertensive disorder characterized by maternal endothelial activation and increased ratio of soluble fms-like tyrosine kinase-1 (sFlt-1) inhibitor to placental growth factor (PlGF). The von Willebrand factor (VWF)/ADAMTS-13 axis is of interest because of the underlying endothelial activation and clinical overlap with pregnancy-associated thrombotic thrombocytopenic purpura.

OBJECTIVES

To assess VWF, ADAMTS-13, and VWF/ADAMTS-13 ratio in preeclampsia and look for associations with sFlt-1/PlGF ratio and clinical features.

METHODS

Thirty-four preeclampsia cases and 48 normal pregnancies were assessed in a case-control study. Twelve normal pregnancies in women with a history of preeclampsia formed an additional comparator group. VWF antigen (VWF:Ag) and VWF activity (VWF:Ac [VWF:glycoprotein IbM]) were measured via automated immunoturbidimetric assay, ADAMTS-13 activity was measured via fluorescence resonance energy transfer-VWF73 assay, and sFlt-1 and PlGF were measured via enzyme-linked immunosorbent assay.

RESULTS

VWF:Ag was higher in preeclampsia than in normal pregnancy (median, 3.07 vs 1.87 IU/mL; P < .0001). ADAMTS-13 activity was slightly lower (median, 89.6 vs 94.4 IU/dL; P = .02), with no severe deficiencies. Significant elevations in VWF:Ac were not observed in preeclampsia, resulting in reduced VWF:Ac/VWF:Ag ratios (median, 0.77 vs 0.97; P < .0001). VWF:Ag/ADAMTS-13 ratios were significantly higher in preeclampsia (median, 3.42 vs 2.06; P < .0001), with an adjusted odds ratio of 19.2 for a ratio of >2.7 (>75th centile of normal pregnancy). Those with a history of preeclampsia had similar ratios to normal pregnant controls. VWF:Ag/ADAMTS-13 and sFlt-1/PlGF were not correlated. However, percentage reduction in platelets correlated positively with VWF:Ac (P = .01), VWF:Ac/VWF:Ag ratio (P = .004), and sFlt-1/PlGF ratio (P = .01).

CONCLUSION

The VWF/ADAMTS-13 axis is significantly altered in preeclampsia. Further investigation of potential clinical utility is warranted.

Human Placenta and Evolving Insights into Pathological Changes of Preeclampsia: A Comprehensive Review of the Last Decade

The placenta, the foremost and multifaceted organ in fetal and maternal biology, is pivotal in facilitating optimal intrauterine fetal development. Remarkably, despite its paramount significance, the placenta remains enigmatic, meriting greater comprehension given its central influence on the health trajectories of both the fetus and the mother. Preeclampsia (PE) and intrauterine fetal growth restriction (IUGR), prevailing disorders of pregnancy, stem from compromised placental development. PE, characterized by heightened mortality and morbidity risks, afflicts 5-7% of global pregnancies, its etiology shrouded in ambiguity. Pertinent pathogenic hallmarks of PE encompass inadequate restructuring of uteroplacental spiral arteries, placental ischemia, and elevated levels of vascular endothelial growth factor receptor-1 (VEGFR-1), also recognized as soluble FMS-like tyrosine kinase-1 (sFlt-1). During gestation, the placental derivation of sFlt-1 accentuates its role as an inhibitory receptor binding to VEGF-A and placental growth factor (PlGF), curtailing target cell accessibility. This review expounds upon the placenta's defining cellular component of the trophoblast, elucidates the intricacies of PE pathogenesis, underscores the pivotal contribution of sFlt-1 to maternal pathology and fetal safeguarding, and surveys recent therapeutic strides witnessed in the past decade.

An Interpretable Longitudinal Preeclampsia Risk Prediction Using Machine Learning

Background

Preeclampsia is a pregnancy-specific disease characterized by new onset hypertension after 20 weeks of gestation that affects 2-8% of all pregnancies and contributes to up to 26% of maternal deaths. Despite extensive clinical research, current predictive tools fail to identify up to 66% of patients who will develop preeclampsia. We sought to develop a tool to longitudinally predict preeclampsia risk.

Methods

In this retrospective model development and validation study, we examined a large cohort of patients who delivered at six community and two tertiary care hospitals in the New England region between 02/2015 and 06/2023. We used sociodemographic, clinical diagnoses, family history, laboratory, and vital signs data. We developed eight datasets at 14, 20, 24, 28, 32, 36, 39 weeks gestation and at the hospital admission for delivery. We created linear regression, random forest, xgboost, and deep neural networks to develop multiple models and compared their performance. We used Shapley values to investigate the global and local explainability of the models and the relationships between the predictive variables.

Findings

Our study population (N=120,752) had an incidence of preeclampsia of 5.7% (N=6,920). The performance of the models as measured using the area under the curve, AUC, was in the range 0.73-0.91, which was externally validated. The relationships between some of the variables were complex and non-linear; in addition, the relative significance of the predictors varied over the pregnancy. Compared to the current standard of care for preeclampsia risk stratification in the first trimester, our model would allow 48.6% more at-risk patients to be identified.

Interpretation

Our novel preeclampsia prediction tool would allow clinicians to identify patients at risk early and provide personalized predictions, as well as longitudinal predictions throughout pregnancy.

Funding

National Institutes of Health, Anesthesia Patient Safety Foundation.

RESEARCH IN CONTEXT

Evidence before this study: Current tools for the prediction of preeclampsia are lacking as they fail to identify up to 66% of the patients who develop preeclampsia. We searched PubMed, MEDLINE, and the Web of Science from database inception to May 1, 2023, using the keywords "deep learning", "machine learning", "preeclampsia", "artificial intelligence", "pregnancy complications", and "predictive models". We identified 13 studies that employed machine learning to develop prediction models for preeclampsia risk based on clinical variables. Among these studies, six included biomarkers such as serum placental growth factor, pregnancy-associated plasma protein A, and uterine artery pulsatility index, which are not routinely available in our clinical practice; two studies were in diverse cohorts of more than 100 000 patients, and two studies developed longitudinal predictions using medical records data. However, most studies have limited depth, concerns about data leakage, overfitting, or lack of generalizability.Added value of this study: We developed a comprehensive longitudinal predictive tool based on routine clinical data that can be used throughout pregnancy to predict the risk of preeclampsia. We tested multiple types of predictive models, including machine learning and deep learning models, and demonstrated high predictive power. We investigated the changes over different time points of individual and group variables and found previously known and novel relationships between variables such as red blood cell count and preeclampsia risk.Implications of all the available evidence: Longitudinal prediction of preeclampsia using machine learning can be achieved with high performance. Implementation of an accurate predictive tool within the electronic health records can aid clinical care and identify patients at heightened risk who would benefit from aspirin prophylaxis, increased surveillance, early diagnosis, and escalation in care. These results highlight the potential of using artificial intelligence in clinical decision support, with the ultimate goal of reducing iatrogenic preterm birth and improving perinatal care.

Maternal Serum Activin A, Inhibin A and Follistatin-Related Proteins across Preeclampsia: Insights into Their Role in Pathogenesis and Prediction

BACKGROUND

Within the endocrine-paracrine signalling network at the maternal-foetal interface, the activin-inhibin-follistatin system modulates extravillous trophoblast invasion, suggesting a potential role in preeclampsia pathogenesis. This study aimed to compile the evidence published in the last decade regarding the variation in maternal serum activins, inhibin- and follistatin-related proteins in preeclamptic pregnancies compared to healthy pregnancies, and to discuss their role in predicting and understanding the pathophysiology of preeclampsia.

MATERIAL AND METHODS

A scoping review was conducted in MEDLINE, EMBASE and LILACS databases to identify studies published within the last ten years (2012-2022).

RESULTS

Thirty studies were included. None of the studies addressed maternal serum changes of isoforms different from activin A, inhibin A, follistatin, and follistatin-like 3. Sixteen studies evaluated the potential of these isoforms in predicting preeclampsia through the area under the curve from a receiver operating characteristic curve.

CONCLUSIONS

In preeclampsia, inhibin A is upregulated in all trimesters, whereas activin A increases exclusively in the late second and third trimesters. Serum follistatin levels are reduced in women with preeclampsia during the late second and third trimesters. However, changes in follistatin-like 3 remain inconclusive. Inhibin A and activin A can potentially serve as biomarkers of early-onset preeclampsia based on the outcomes of the receiver operating characteristic curve analysis. Further investigations are encouraged to explore the feasibility of quantifying maternal serum levels of activin A and inhibin A as a clinical tool in early preeclampsia prediction.

Preeclampsia at term can be classified into 2 clusters with different clinical characteristics and outcomes based on angiogenic biomarkers in maternal blood

BACKGROUND

An antiangiogenic state has emerged as a mechanism of disease in preeclampsia. Angiogenic biomarkers are used in the risk assessment of this syndrome, particularly of early disease. The role of an antiangiogenic state in late preeclampsia is unclear.

OBJECTIVE

This study aimed to determine the prevalence, characteristics, and clinical significance of angiogenic/antiangiogenic factor abnormalities in women with preeclampsia stratified according to gestational age at delivery.

STUDY DESIGN

Two studies were conducted: (1) a longitudinal nested case-control study comprising women with preeclampsia (n=151) and a control group (n=540); and (2) a case series of patients with preeclampsia (n=452). In patients with preeclampsia, blood was collected at the time of diagnosis. Plasma concentrations of placental growth factor and soluble fms-like tyrosine kinase-1 were determined by enzyme-linked immunosorbent assays. An abnormal angiogenic profile was defined as a plasma ratio of placental growth factor and soluble fms-like tyrosine kinase-1 expressed as a multiple of the median <10th percentile for gestational age based on values derived from the longitudinal study. The proportion of patients diagnosed with preeclampsia who had an abnormal angiogenic profile was determined in the case-series participants and stratified by gestational age at delivery into early (≤34 weeks), intermediate (34.1-36.9 weeks), and term (≥37 weeks) preeclampsia. The demographics, clinical characteristics, and pregnancy outcomes of women with preeclampsia with and without an abnormal angiogenic profile were compared.

RESULTS

The prevalence of an abnormal angiogenic profile was higher in preterm than in term preeclampsia (for early, intermediate, and term in the case-control study: 90%, 100%, and 39%; for the case series: 98%, 80%, and 55%, respectively). Women with preeclampsia at term who had an abnormal angiogenic profile were more frequently nulliparous (57% vs 35%), less likely to smoke (14% vs 26%), at greater risk for maternal (14% vs 5%) or neonatal (7% vs 1%) complications, and more often had placental lesions consistent with maternal vascular malperfusion (42% vs 23%; all, P<.05) than those without an abnormal profile. Women with preeclampsia at term who had a normal angiogenic profile had a higher frequency of chronic hypertension (36% vs 21%) and were more likely to have class ≥2 obesity (41% vs 23%) than those with an abnormal profile (both, P<.05).

CONCLUSION

Patients with early preeclampsia had an abnormal angiogenic profile in virtually all cases, whereas only 50% of women with preeclampsia at term had such abnormalities. The profile of angiogenic biomarkers can be used to classify patients with preeclampsia at term, on the basis of mechanisms of disease, into 2 clusters, which have different demographics, clinical characteristics, and risks of adverse maternal and neonatal outcomes. These findings provide a simple approach to classify preeclampsia at term and have implications for future clinical care and research.

A Comprehensive and Bias-Free Machine Learning Approach for Risk Prediction of Preeclampsia with Severe Features in a Nulliparous Study Cohort

Objective

Preeclampsia is one of the leading causes of maternal morbidity, with consequences during and after pregnancy. Because of its diverse clinical presentation, preeclampsia is an adverse pregnancy outcome that is uniquely challenging to predict and manage. In this paper, we developed machine learning models that predict the onset of preeclampsia with severe features or eclampsia at discrete time points in a nulliparous pregnant study cohort.

Materials and Methods

The prospective study cohort to which we applied machine learning is the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-be (nuMoM2b) study, which contains information from eight clinical sites across the US. Maternal serum samples were collected for 1,857 individuals between the first and second trimesters. These patients with serum samples collected are selected as the final cohort.

Results

Our prediction models achieved an AUROC of 0.72 (95% CI, 0.69-0.76), 0.75 (95% CI, 0.71-0.79), and 0.77 (95% CI, 0.74-0.80), respectively, for the three visits. Our initial models were biased toward non-Hispanic black participants with a high predictive equality ratio of 1.31. We corrected this bias and reduced this ratio to 1.14. The top features stress the importance of using several tests, particularly for biomarkers and ultrasound measurements. Placental analytes were strong predictors for screening for the early onset of preeclampsia with severe features in the first two trimesters.

Conclusion

Experiments suggest that it is possible to create racial bias-free early screening models to predict the patients at risk of developing preeclampsia with severe features or eclampsia nulliparous pregnant study cohort.

Diagnostic Accuracy of Gestosis Score in Comparison to multi-marker Screening as a Predictor of Preeclampsia at 11-14 Weeks of Pregnancy: A Cohort Study

BACKGROUND

Pre-eclampsia is a pregnancy-specific multisystemic disorder associated with adverse feto-maternal outcomes. Low-dose Aspirin therapy started in early pregnancy in high-risk women, has significantly reduced the chances of developing PE. Therefore, screening and identification of at-risk mothers are crucial. The present study was planned to study the predictive ability of gestosis score in predicting early-onset pre-eclampsia by comparing it with the multi-marker model.

MATERIAL AND METHODS

One hundred sixteen women, more than 19 years of age, with live singleton pregnancy at 11-13 weeks of gestation were recruited from the antenatal outpatient department and formed the study cohort. After a detailed history, screening for pre-eclampsia was performed both by multi-marker screening and by gestosis score. Diagnostic accuracy was compared for the two methods of screening.

RESULTS

The incidence of pre-eclampsia in the present study cohort was 26.7%. The sensitivity of gestosis score >/= 3 was 84.38% (67.21-94.72) and specificity was 93.18% (85.75-97.46 %). The positive predictive value was 81.82% (67.2%-90.81%), and the negative predictive value was 94.25 (87.98 - 97.35%). The diagnostic accuracy of the gestosis score was 90.83%.

CONCLUSION

Gestosis scoring is a potential tool that can be used as a cost-effective screening method for pre-eclampsia at 11-14 weeks of gestation in low-resource settings. The sensitivity and negative predictive value of the gestosis score is comparable to multi-marker screening using maternal factors, MAP, Uterine artery PI, PAPP-A, and PlGF.

Comparison of machine learning and logistic regression as predictive models for adverse maternal and neonatal outcomes of preeclampsia: A retrospective study

Introduction

Preeclampsia, one of the leading causes of maternal and fetal morbidity and mortality, demands accurate predictive models for the lack of effective treatment. Predictive models based on machine learning algorithms demonstrate promising potential, while there is a controversial discussion about whether machine learning methods should be recommended preferably, compared to traditional statistical models.

Methods

We employed both logistic regression and six machine learning methods as binary predictive models for a dataset containing 733 women diagnosed with preeclampsia. Participants were grouped by four different pregnancy outcomes. After the imputation of missing values, statistical description and comparison were conducted preliminarily to explore the characteristics of documented 73 variables. Sequentially, correlation analysis and feature selection were performed as preprocessing steps to filter contributing variables for developing models. The models were evaluated by multiple criteria.

Results

We first figured out that the influential variables screened by preprocessing steps did not overlap with those determined by statistical differences. Secondly, the most accurate imputation method is K-Nearest Neighbor, and the imputation process did not affect the performance of the developed models much. Finally, the performance of models was investigated. The random forest classifier, multi-layer perceptron, and support vector machine demonstrated better discriminative power for prediction evaluated by the area under the receiver operating characteristic curve, while the decision tree classifier, random forest, and logistic regression yielded better calibration ability verified, as by the calibration curve.

Conclusion

Machine learning algorithms can accomplish prediction modeling and demonstrate superior discrimination, while Logistic Regression can be calibrated well. Statistical analysis and machine learning are two scientific domains sharing similar themes. The predictive abilities of such developed models vary according to the characteristics of datasets, which still need larger sample sizes and more influential predictors to accumulate evidence.