Machine-learning-based prediction of pre-eclampsia using first-trimester maternal characteristics and biomarkers

First-trimester preterm preeclampsia prediction model for prevention with low-dose aspirin

INTRODUCTION

Preeclampsia (PE) is a major maternal and fetal threat. Previous risk-scoring methods in guidelines lacked precision. The Fetal Medicine Foundation (FMF) proposed a first-trimester PE screening model using Bayes' theorem.

PE PREDICTION MODEL

FMF prediction model combines maternal characteristics and medical/obstetrical history to determine prior risk and further incorporate maternal blood pressure, maternal serum biomarkers, and uterine Doppler pulsatility index expressed as multiples of the median (MoM) to estimate posterior risk.

LOW-DOSE ASPIRIN PREVENTION

Low-dose aspirin is one of the potential PE prevention strategies. Initiating it before 16 weeks is crucial. Aspirin's antiplatelet and anti-inflammatory properties align with PE's pathophysiology. Dosing and resistance warrant further study, but a standard regimen of 150 mg nightly, starting before 16 weeks, is widely supported.

PE PREVENTION IN PRACTICE

Clinical trials, including ASPRE, affirm aspirin's role in PE prevention. Starting aspirin based on FMF screening significantly reduces preterm PE and associated complications.

ADVANCEMENTS AND PROSPECTS

Emerging research explores predictors like maternal ophthalmic arterial waveform. Regional variations, especially in Asian populations, are considered. Machine learning and AI show promise, but examiner expertise remains essential for accurate prediction. In conclusion, integrating FMF's first-trimester PE screening with low-dose aspirin offers a promising strategy. Further advancements may enhance precision and broaden prevention efforts.

Performance of machine-learning approach for prediction of pre-eclampsia in a middle-income country

OBJECTIVE

Pre-eclampsia (PE) is a serious complication of pregnancy associated with maternal and fetal morbidity and mortality. As current prediction models have limitations and may not be applicable in resource-limited settings, we aimed to develop a machine-learning (ML) algorithm that offers a potential solution for developing accurate and efficient first-trimester prediction of PE.

METHODS

We conducted a prospective cohort study in Mexico City, Mexico to develop a first-trimester prediction model for preterm PE (pPE) using ML. Maternal characteristics and locally derived multiples of the median (MoM) values for mean arterial pressure, uterine artery pulsatility index and serum placental growth factor were used for variable selection. The dataset was split into training, validation and test sets. An elastic-net method was employed for predictor selection, and model performance was evaluated using area under the receiver-operating-characteristics curve (AUC) and detection rates (DR) at 10% false-positive rates (FPR).

RESULTS

The final analysis included 3050 pregnant women, of whom 124 (4.07%) developed PE. The ML model showed good performance, with AUCs of 0.897, 0.963 and 0.778 for pPE, early-onset PE (ePE) and any type of PE (all-PE), respectively. The DRs at 10% FPR were 76.5%, 88.2% and 50.1% for pPE, ePE and all-PE, respectively.

CONCLUSIONS

Our ML model demonstrated high accuracy in predicting pPE and ePE using first-trimester maternal characteristics and locally derived MoM. The model may provide an efficient and accessible tool for early prediction of PE, facilitating timely intervention and improved maternal and fetal outcome. © 2023 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

Validation of machine-learning model for first-trimester prediction of pre-eclampsia using cohort from PREVAL study

OBJECTIVE

Effective first-trimester screening for pre-eclampsia (PE) can be achieved using a competing-risks model that combines risk factors from the maternal history with multiples of the median (MoM) values of biomarkers. A new model using artificial intelligence through machine-learning methods has been shown to achieve similar screening performance without the need for conversion of raw data of biomarkers into MoM. This study aimed to investigate whether this model can be used across populations without specific adaptations.

METHODS

Previously, a machine-learning model derived with the use of a fully connected neural network for first-trimester prediction of early (< 34 weeks), preterm (< 37 weeks) and all PE was developed and tested in a cohort of pregnant women in the UK. The model was based on maternal risk factors and mean arterial blood pressure (MAP), uterine artery pulsatility index (UtA-PI), placental growth factor (PlGF) and pregnancy-associated plasma protein-A (PAPP-A). In this study, the model was applied to a dataset of 10 110 singleton pregnancies examined in Spain who participated in the first-trimester PE validation (PREVAL) study, in which first-trimester screening for PE was carried out using the Fetal Medicine Foundation (FMF) competing-risks model. The performance of screening was assessed by examining the area under the receiver-operating-characteristics curve (AUC) and detection rate (DR) at a 10% screen-positive rate (SPR). These indices were compared with those derived from the application of the FMF competing-risks model. The performance of screening was poor if no adjustment was made for the analyzer used to measure PlGF, which was different in the UK and Spain. Therefore, adjustment for the analyzer used was performed using simple linear regression.

RESULTS

The DRs at 10% SPR for early, preterm and all PE with the machine-learning model were 84.4% (95% CI, 67.2-94.7%), 77.8% (95% CI, 66.4-86.7%) and 55.7% (95% CI, 49.0-62.2%), respectively, with the corresponding AUCs of 0.920 (95% CI, 0.864-0.975), 0.913 (95% CI, 0.882-0.944) and 0.846 (95% CI, 0.820-0.872). This performance was achieved with the use of three of the biomarkers (MAP, UtA-PI and PlGF); inclusion of PAPP-A did not provide significant improvement in DR. The machine-learning model had similar performance to that achieved by the FMF competing-risks model (DR at 10% SPR, 82.7% (95% CI, 69.6-95.8%) for early PE, 72.7% (95% CI, 62.9-82.6%) for preterm PE and 55.1% (95% CI, 48.8-61.4%) for all PE) without requiring specific adaptations to the population.

CONCLUSIONS

A machine-learning model for first-trimester prediction of PE based on a neural network provides effective screening for PE that can be applied in different populations. However, before doing so, it is essential to make adjustments for the analyzer used for biochemical testing. © 2023 International Society of Ultrasound in Obstetrics and Gynecology.

Preterm preeclampsia screening using biomarkers: combining phenotypic classifiers into robust prediction models

BACKGROUND

Preeclampsia screening is a critical component of antenatal care worldwide. Currently, the most developed screening test for preeclampsia at 11 to 13 weeks' gestation integrates maternal demographic characteristics and medical history with 3 biomarkers-serum placental growth factor, mean arterial pressure, and uterine artery pulsatility index-to identify approximately 75% of women who develop preterm preeclampsia with delivery before 37 weeks of gestation. It is generally accepted that further improvements to preeclampsia screening require the use of additional biomarkers. We recently reported that the levels of specific metabolites and metabolite ratios are associated with preterm preeclampsia. Notably, for several of these markers, preterm preeclampsia prediction varied according to maternal body mass index class. These findings motivated us to study whether patient classification allowed for combining metabolites with the current biomarkers more effectively to improve prediction of preterm preeclampsia.

OBJECTIVE

This study aimed to investigate whether metabolite biomarkers can improve biomarker-based preterm preeclampsia prediction in 3 screening resource scenarios according to the availability of: (1) placental growth factor, (2) placental growth factor+mean arterial pressure, and (3) placental growth factor+mean arterial pressure+uterine artery pulsatility index.

STUDY DESIGN

This was an observational case-control study, drawn from a large prospective screening study at 11 to 13 weeks' gestation on the prediction of pregnancy complications, conducted at King's College Hospital, London, United Kingdom. Maternal blood samples were also collected for subsequent research studies. We used liquid chromatography-mass spectrometry to quantify levels of 50 metabolites previously associated with pregnancy complications in plasma samples from singleton pregnancies. Biomarker data, normalized using multiples of medians, on 1635 control and 106 preterm preeclampsia pregnancies were available for model development. Modeling was performed using a methodology that generated a prediction model for preterm preeclampsia in 4 consecutive steps: (1) z-normalization of predictors, (2) combinatorial modeling of so-called (weak) classifiers in the unstratified patient set and in discrete patient strata based on body mass index and/or race, (3) selection of classifiers, and (4) aggregation of the selected classifiers (ie, bagging) into the final prediction model. The prediction performance of models was evaluated using the area under the receiver operating characteristic curve, and detection rate at 10% false-positive rate.

RESULTS

First, the predictor development methodology itself was evaluated. The patient set was split into a training set (2/3) and a test set (1/3) for predictor model development and internal validation. A prediction model was developed for each of the 3 different predictor panels, that is, placental growth factor+metabolites, placental growth factor+mean arterial pressure+metabolites, and placental growth factor+mean arterial pressure+uterine artery pulsatility index+metabolites. For all 3 models, the area under the receiver operating characteristic curve in the test set did not differ significantly from that of the training set. Next, a prediction model was developed using the complete data set for the 3 predictor panels. Among the 50 metabolites available for modeling, 26 were selected across the 3 prediction models; 21 contributed to at least 2 out of the 3 prediction models developed. Each time, area under the receiver operating characteristic curve and detection rate were significantly higher with the new prediction model than with the reference model. Markedly, the estimated detection rate with the placental growth factor+mean arterial pressure+metabolites prediction model in all patients was 0.58 (95% confidence interval, 0.49-0.70), a 15% increase (P<.001) over the detection rate of 0.43 (95% confidence interval, 0.33-0.55) estimated for the reference placental growth factor+mean arterial pressure. The same prediction model significantly improved detection in Black (14%) and White (19%) patients, and in the normal-weight group (18.5≤body mass index<25) and the obese group (body mass index≥30), with respectively 19% and 20% more cases detected, but not in the overweight group, when compared with the reference model. Similar improvement patterns in detection rates were found in the other 2 scenarios, but with smaller improvement amplitudes.

CONCLUSION

Metabolite biomarkers can be combined with the established biomarkers of placental growth factor, mean arterial pressure, and uterine artery pulsatility index to improve the biomarker component of early-pregnancy preterm preeclampsia prediction tests. Classification of the pregnant women according to the maternal characteristics of body mass index and/or race proved instrumental in achieving improved prediction. This suggests that maternal phenotyping can have a role in improving the prediction of obstetrical syndromes such as preeclampsia.

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.

Potential biomarkers for late-onset and term preeclampsia: A scoping review

Preeclampsia is a progressive, multisystem pregnancy disorder. According to the time of onset or delivery, preeclampsia has been subclassified into early-onset (&lt;34 weeks) and late-onset (≥34 weeks), or preterm (&lt;37 weeks) and term (≥37 weeks). Preterm preeclampsia can be effectively predicted at 11–13 weeks well before onset, and its incidence can be reduced by preventively using low-dose aspirin. However, late-onset and term preeclampsia are more prevalent than early forms and still lack effective predictive and preventive measures. This scoping review aims to systematically identify the evidence of predictive biomarkers reported in late-onset and term preeclampsia. This study was conducted based on the guidance of the Joanna Briggs Institute (JBI) methodology for scoping reviews. The Preferred Reporting Items for Systematic Reviews and Meta-Analysis extension for scoping reviews (PRISMA-ScR) was used to guide the study. The following databases were searched for related studies: PubMed, Web of Science, Scopus, and ProQuest. Search terms contain “preeclampsia,” “late-onset,” “term,” “biomarker,” or “marker,” and other synonyms combined as appropriate using the Boolean operators “AND” and “OR.” The search was restricted to articles published in English from 2012 to August 2022. Publications were selected if study participants were pregnant women and biomarkers were detected in maternal blood or urine samples before late-onset or term preeclampsia diagnosis. The search retrieved 4,257 records, of which 125 studies were included in the final assessment. The results demonstrate that no single molecular biomarker presents sufficient clinical sensitivity and specificity for screening late-onset and term preeclampsia. Multivariable models combining maternal risk factors with biochemical and/or biophysical markers generate higher detection rates, but they need more effective biomarkers and validation data for clinical utility. This review proposes that further research into novel biomarkers for late-onset and term preeclampsia is warranted and important to find strategies to predict this complication. Other critical factors to help identify candidate markers should be considered, such as a consensus on defining preeclampsia subtypes, optimal testing time, and sample types.