Machine Learning Models for Predicting Adverse Pregnancy Outcomes in Pregnant Women with Systemic Lupus Erythematosus.

Predicting adverse outcomes is essential for pregnant women with systemic lupus erythematosus (SLE) to minimize risks. Applying statistical analysis may be limited for the small sample size of childbearing patients, while the informative medical records could be provided. This study aimed to develop predictive models applying machine learning (ML) techniques to explore more information. We performed a retrospective analysis of 51 pregnant women exhibiting SLE, including 288 variables. After correlation analysis and feature selection, six ML models were applied to the filtered dataset. The efficiency of these overall models was evaluated by the Receiver Operating Characteristic Curve. Meanwhile, real-time models with different timespans based on gestation were also explored. Eighteen variables demonstrated statistical differences between the two groups; more than forty variables were screened out by ML variable selection strategies as contributing predictors, while the overlap of variables were the influential indicators testified by the two selection strategies. The Random Forest (RF) algorithm demonstrated the best discrimination ability under the current dataset for overall predictive models regardless of the data missing rate, while Multi-Layer Perceptron models ranked second. Meanwhile, RF achieved best performance when assessing the real-time predictive accuracy of models. ML models could compensate the limitation of statistical methods when the small sample size problem happens along with numerous variables acquired, while RF classifier performed relatively best when applied to such structured medical records.

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.

A bibliometric analysis of the research on preeclampsia in the first two decades of the twenty-first century.

BACKGROUND: Preeclampsia still remains one of the leading causes of maternal and perinatal mortality worldwide. Despite the concerted efforts of researchers, only a little improvement has been seen. Clinical decision-making is based on the published literatures. With the explosive growth of medical documents in recent decades, a bibliometric method is essential for assessing the intellectual contributions, major components and potential trends. METHODS: Web of Science Core Collections was selected as the original database and datasets were retrieved consisting of literatures published from 2000 to 2020. Different bibliometric software were employed to visualize the co-authorship network, citation analysis and research theme detection. RESULTS: A total of 25497 articles and 3668 reviews were obtained. Despite the number of publications increased annually, the quantity of high-quality contributions did not elevate accordingly. Clinical practitioners should be alerted to the false bloom of achievements and the yield of improvement in future research. Nicolaides Kypros H was found to be the most productive and influential researcher. University of Pittsburgh was the most productive institution whereas Harvard University showed its leading academic status. America located at the central point in global collaboration and scholarship network. Reference citation analysis revealed the top landmark articles. Moreover, keywords co-occurrence analysis and burst detection certificated the lack of novel themes in this field, which needs further efforts. CONCLUSION: This study provides the overall landscape of science mapping in recent two decades in the field of preeclampsia, with the aim of identifying evolution of research topics and promoting potential concentration or collaboration in the future.

Corrigendum: Long Non-Coding RNA Gas5 is Associated with Preeclampsia and Regulates Biological Behaviors of Trophoblast via MicroRNA-21.

[This corrects the article DOI: 10.3389/fgene.2020.00188.].

Long Non-coding RNA Gas5 Is Associated With Preeclampsia and Regulates Biological Behaviors of Trophoblast via MicroRNA-21.

Preeclampsia is a lethal pregnancy specific hypertensive disorder involving multisystem. Despite extensive studies to investigate the causes of preeclampsia, the pathogenesis still remains largely unknown. Long non-coding RNAs (lncRNAs) are a diverse class of non-translated RNAs which play a crucial part in various biological phenomena. Although lncRNA Growth Arrest-Specific 5 (GAS5) aberrantly expressed in multiple cancer tissues and is implicated in multiple biological processes of tumor cells, little is known about its role in preeclampsia. In this study, 40 patients with preeclampsia and 32 gestational age matched normotension pregnant women were recruited. Using quantitative real-time polymerase chain reaction (qRT-PCR), we found higher expression of GAS5 in placenta of preclamsia affected women. The level of GAS5 existed strongly in correlation with Thrombin Time indicating coagulation function and other clinical parameters by Pearson correlation analysis. Then we constructed the GAS5 lentivirus expression vectors and transfected into human trophoblast cell lines HTR-8/SVneo and JEG-3. Using in vitro cell culture studies, we found an inhibited effect of GAS5 on proliferative ability, migratory ability and invasive ability however; no effect on apoptosis was detected. Further mechanistic analysis found that GAS5 modulated microRNA-21 (miR-21) in an opposite variation tendency by qRT-PCR and rescue experiment. In addition, inhibition of GAS5 promoted the activation of PI3K/AKT signaling pathway and its downstream proteins covering MMP-9 and TP53 as evident from our qRT-PCR and western blot analyses. Thus, we suggested that GAS5 might involve in pregnancy with preeclampsia by influencing the biological functions of trophoblast cells through the regulation of miR-21 and activation of PI3K/AKT signaling pathway and its downstream targets, which may contribute to reveal the nature of preeclampsia.