Prediction of gestational diabetes based on nationwide electronic health records

Design, rationale and protocol for Glycemic Observation and Metabolic Outcomes in Mothers and Offspring (GO MOMs): an observational cohort study

INTRODUCTION

Given the increasing prevalence of both obesity and pre-diabetes in pregnant adults, there is growing interest in identifying hyperglycaemia in early pregnancy to optimise maternal and perinatal outcomes. Multiple organisations recommend first-trimester diabetes screening for individuals with risk factors; however, the benefits and drawbacks of detecting glucose abnormalities more mild than overt diabetes in early gestation and the best screening method to detect such abnormalities remain unclear.

METHODS AND ANALYSIS

The goal of the Glycemic Observation and Metabolic Outcomes in Mothers and Offspring study (GO MOMs) is to evaluate how early pregnancy glycaemia, measured using continuous glucose monitoring and oral glucose tolerance testing, relates to the diagnosis of gestational diabetes (GDM) at 24-28 weeks' gestation (maternal primary outcome) and large-for-gestational-age birth weight (newborn primary outcome). Secondary objectives include relating early pregnancy glycaemia to other adverse pregnancy outcomes and comprehensively detailing longitudinal changes in glucose over the course of pregnancy. GO MOMs enrolment began in April 2021 and will continue for 3.5 years with a target sample size of 2150 participants.

ETHICS AND DISSEMINATION

GO MOMs is centrally overseen by Vanderbilt University's Institutional Review Board and an Observational Study Monitoring Board appointed by National Institute of Diabetes and Digestive and Kidney Diseases. GO MOMs has potential to yield data that will improve understanding of hyperglycaemia in pregnancy, elucidate better approaches for early pregnancy GDM screening, and inform future clinical trials of early GDM treatment.

TRIAL REGISTRATION NUMBER

NCT04860336.

A Theoretical Exploration of Artificial Intelligence's Impact on Feto-Maternal Health from Conception to Delivery

The implementation of Artificial Intelligence (AI) in healthcare is enhancing diagnostic accuracy in clinical setups. The use of AI in healthcare is steadily increasing with advancing technology, extending beyond disease diagnosis to encompass roles in feto-maternal health. AI harnesses Machine Learning (ML), Natural Language Processing (NLP), Artificial Neural Networks (ANN), and computer vision to analyze data and draw conclusions. Considering maternal health, ML analyzes vast datasets to predict maternal and fetal health outcomes, while NLP interprets medical texts and patient records to assist in diagnosis and treatment decisions. ANN models identify patterns in complex feto-maternal medical data, aiding in risk assessment and intervention planning whereas, computer vision enables the analysis of medical images for early detection of feto-maternal complications. AI facilitates early pregnancy detection, genetic screening, and continuous monitoring of maternal health parameters, providing real-time alerts for deviations, while also playing a crucial role in the early detection of fetal abnormalities through enhanced ultrasound imaging, contributing to informed decision-making. This review investigates into the application of AI, particularly through predictive models, in addressing the monitoring of feto-maternal health. Additionally, it examines potential future directions and challenges associated with these applications.

Machine Learning Approach to Metabolomic Data Predicts Type 2 Diabetes Mellitus Incidence

Metabolomics, with its wealth of data, offers a valuable avenue for enhancing predictions and decision-making in diabetes. This observational study aimed to leverage machine learning (ML) algorithms to predict the 4-year risk of developing type 2 diabetes mellitus (T2DM) using targeted quantitative metabolomics data. A cohort of 279 cardiovascular risk patients who underwent coronary angiography and who were initially free of T2DM according to American Diabetes Association (ADA) criteria was analyzed at baseline, including anthropometric data and targeted metabolomics, using liquid chromatography (LC)-mass spectroscopy (MS) and flow injection analysis (FIA)-MS, respectively. All patients were followed for four years. During this time, 11.5% of the patients developed T2DM. After data preprocessing, 362 variables were used for ML, employing the Caret package in R. The dataset was divided into training and test sets (75:25 ratio) and we used an oversampling approach to address the classifier imbalance of T2DM incidence. After an additional recursive feature elimination step, identifying a set of 77 variables that were the most valuable for model generation, a Support Vector Machine (SVM) model with a linear kernel demonstrated the most promising predictive capabilities, exhibiting an F1 score of 50%, a specificity of 93%, and balanced and unbalanced accuracies of 72% and 88%, respectively. The top-ranked features were bile acids, ceramides, amino acids, and hexoses, whereas anthropometric features such as age, sex, waist circumference, or body mass index had no contribution. In conclusion, ML analysis of metabolomics data is a promising tool for identifying individuals at risk of developing T2DM and opens avenues for personalized and early intervention strategies.

The Mediating Role of Physical Activity Self-Efficacy in Predicting Moderate-Intensity Physical Activity in Pregnant People at High Risk for Gestational Diabetes

INTRODUCTION

Gestational diabetes mellitus (GDM) is a common medical complication in pregnancy. Moderate-intensity physical activity during pregnancy can lower the risk of GDM. However, the relationship between moderate-intensity physical activity and correlated factors among pregnant people at high risk for GDM remains unknown.

METHODS

A cross-sectional study was conducted in China. Two hundred fifty-two participants completed the Pregnancy Physical Activity Questionnaire, Pregnancy Physical Activity Self-Efficacy Scale, Physical Activity Knowledge Questionnaire, Physical Activity Social Support Scale, 7-item Generalized Anxiety Disorder Scale, Edinburgh Postnatal Depression Scale, and a sociodemographic data sheet. Structural equation modeling was used to explore the direct and indirect associations between the study variables.

RESULTS

A total of 51.6% of the participants did not meet the current physical activity guidelines. Only physical activity self-efficacy was significantly correlated with moderate-intensity physical activity. Physical activity self-efficacy mediated the relationship between moderate-intensity physical activity and knowledge of physical activity, social support for physical activity, and anxiety symptoms. Furthermore, knowledge of physical activity was also associated with improved moderate-intensity physical activity mediated by reduced anxiety symptoms and increased physical activity self-efficacy.

CONCLUSION

Our study revealed a high prevalence of not meeting current physical activity guidelines among pregnant people at high risk for GDM. Physical activity self-efficacy played an important mediating role in predicting moderate-intensity physical activity. Future studies should focus on enhancing self-efficacy to improve moderate-intensity physical activity for pregnant people at high risk for GDM.

Early prediction of gestational diabetes mellitus using maternal demographic and clinical risk factors

OBJECTIVE

To build and validate an early risk prediction model for gestational diabetes mellitus (GDM) based on first-trimester electronic medical records including maternal demographic and clinical risk factors.

METHODS

To develop and validate a GDM prediction model, two datasets were used in this retrospective study. One included data of 14,015 pregnant women from Máxima Medical Center (MMC) in the Netherlands. The other was from an open-source database nuMoM2b including data of 10,038 nulliparous pregnant women, collected in the USA. Widely used maternal demographic and clinical risk factors were considered for modeling. A GDM prediction model based on elastic net logistic regression was trained from a subset of the MMC data. Internal validation was performed on the remaining MMC data to evaluate the model performance. For external validation, the prediction model was tested on an external test set from the nuMoM2b dataset.

RESULTS

An area under the receiver-operating-characteristic curve (AUC) of 0.81 was achieved for early prediction of GDM on the MMC test data, comparable to the performance reported in previous studies. While the performance markedly decreased to an AUC of 0.69 when testing the MMC-based model on the external nuMoM2b test data, close to the performance trained and tested on the nuMoM2b dataset only (AUC = 0.70).

Healthcare on the brink: navigating the challenges of an aging society in the United States

The US healthcare system is at a crossroads. With an aging population requiring more care and a strained system facing workforce shortages, capacity issues, and fragmentation, innovative solutions and policy reforms are needed. This paper aims to spark dialogue and collaboration among healthcare stakeholders and inspire action to meet the needs of the aging population. Through a comprehensive analysis of the impact of an aging society, this work highlights the urgency of addressing this issue and the importance of restructuring the healthcare system to be more efficient, equitable, and responsive.

Ensemble machine learning reveals key features for diabetes duration from electronic health records

Diabetes is a metabolic disorder that affects more than 420 million of people worldwide, and it is caused by the presence of a high level of sugar in blood for a long period. Diabetes can have serious long-term health consequences, such as cardiovascular diseases, strokes, chronic kidney diseases, foot ulcers, retinopathy, and others. Even if common, this disease is uneasy to spot, because it often comes with no symptoms. Especially for diabetes type 2, that happens mainly in the adults, knowing how long the diabetes has been present for a patient can have a strong impact on the treatment they can receive. This information, although pivotal, might be absent: for some patients, in fact, the year when they received the diabetes diagnosis might be well-known, but the year of the disease unset might be unknown. In this context, machine learning applied to electronic health records can be an effective tool to predict the past duration of diabetes for a patient. In this study, we applied a regression analysis based on several computational intelligence methods to a dataset of electronic health records of 73 patients with diabetes type 1 with 20 variables and another dataset of records of 400 patients of diabetes type 2 with 49 variables. Among the algorithms applied, Random Forests was able to outperform the other ones and to efficiently predict diabetes duration for both the cohorts, with the regression performances measured through the coefficient of determination R2. Afterwards, we applied the same method for feature ranking, and we detected the most relevant factors of the clinical records correlated with past diabetes duration: age, insulin intake, and body-mass index. Our study discoveries can have profound impact on clinical practice: when the information about the duration of diabetes of patient is missing, medical doctors can use our tool and focus on age, insulin intake, and body-mass index to infer this important aspect. Regarding limitations, unfortunately we were unable to find additional dataset of EHRs of patients with diabetes having the same variables of the two analyzed here, so we could not verify our findings on a validation cohort.

Development and validation of machine learning-augmented algorithm for insulin sensitivity assessment in the community and primary care settings: a population-based study in China

Objective

Insulin plays a central role in the regulation of energy and glucose homeostasis, and insulin resistance (IR) is widely considered as the "common soil" of a cluster of cardiometabolic disorders. Assessment of insulin sensitivity is very important in preventing and treating IR-related disease. This study aims to develop and validate machine learning (ML)-augmented algorithms for insulin sensitivity assessment in the community and primary care settings.

Methods

We analyzed the data of 9358 participants over 40 years old who participated in the population-based cohort of the Hubei center of the REACTION study (Risk Evaluation of Cancers in Chinese Diabetic Individuals). Three non-ensemble algorithms and four ensemble algorithms were used to develop the models with 70 non-laboratory variables for the community and 87 (70 non-laboratory and 17 laboratory) variables for the primary care settings to screen the classifier of the state-of-the-art. The models with the best performance were further streamlined using top-ranked 5, 8, 10, 13, 15, and 20 features. Performances of these ML models were evaluated using the area under the receiver operating characteristic curve (AUROC), the area under the precision-recall curve (AUPR), and the Brier score. The Shapley additive explanation (SHAP) analysis was employed to evaluate the importance of features and interpret the models.

Results

The LightGBM models developed for the community (AUROC 0.794, AUPR 0.575, Brier score 0.145) and primary care settings (AUROC 0.867, AUPR 0.705, Brier score 0.119) achieved higher performance than the models constructed by the other six algorithms. The streamlined LightGBM models for the community (AUROC 0.791, AUPR 0.563, Brier score 0.146) and primary care settings (AUROC 0.863, AUPR 0.692, Brier score 0.124) using the 20 top-ranked variables also showed excellent performance. SHAP analysis indicated that the top-ranked features included fasting plasma glucose (FPG), waist circumference (WC), body mass index (BMI), triglycerides (TG), gender, waist-to-height ratio (WHtR), the number of daughters born, resting pulse rate (RPR), etc.

Conclusion

The ML models using the LightGBM algorithm are efficient to predict insulin sensitivity in the community and primary care settings accurately and might potentially become an efficient and practical tool for insulin sensitivity assessment in these settings.

Precise prediction of metabolites patterns using machine learning approaches in distinguishing honey and sugar diets fed to mice

As a natural sweetener produced by honey bees, honey was recognized as being healthier for consumption than table sugar. Our previous study also indicated thatmetaboliteprofiles in mice fed honey and mixedsugardiets aredifferent. However, it is still noteworthy about the batch-to-batch consistency of the metabolic differences between two diet types. Here, the machine learning (ML) algorithms were applied to complement and calibrate HPLC-QTOF/MS-based untargeted metabolomics data. Data were generated from three batches of mice that had the same treatment, which can further mine the metabolite biomarkers. Random Forest and Extra-Trees models could better discriminate between honey and mixed sugar dietary patterns under five-fold cross-validation. Finally, SHapley Additive exPlanations tool identified phosphatidylethanolamine and phosphatidylcholine as reliable metabolic biomarkers to discriminate the honey diet from the mixed sugar diet. This study provides us new ideas for metabolomic analysis of larger data sets.

Differentiation between cerebral alveolar echinococcosis and brain metastases with radiomics combined machine learning approach

BACKGROUND

Cerebral alveolar echinococcosis (CAE) and brain metastases (BM) share similar in locations and imaging appearance. However, they require distinct treatment approaches, with CAE typically treated with chemotherapy and surgery, while BM is managed with radiotherapy and targeted therapy for the primary malignancy. Accurate diagnosis is crucial due to the divergent treatment strategies.

PURPOSE

This study aims to evaluate the effectiveness of radiomics and machine learning techniques based on magnetic resonance imaging (MRI) to differentiate between CAE and BM.

METHODS

We retrospectively analyzed MRI images of 130 patients (30 CAE and 100 BM) from Xinjiang Medical University First Affiliated Hospital and The First People's Hospital of Kashi Prefecture, between January 2014 and December 2022. The dataset was divided into training (91 cases) and testing (39 cases) sets. Three dimensional tumors were segmented by radiologists from contrast-enhanced T1WI images on open resources software 3D Slicer. Features were extracted on Pyradiomics, further feature reduction was carried out using univariate analysis, correlation analysis, and least absolute shrinkage and selection operator (LASSO). Finally, we built five machine learning models, support vector machine, logistic regression, linear discrimination analysis, k-nearest neighbors classifier, and Gaussian naïve bias and evaluated their performance via several metrics including sensitivity (recall), specificity, positive predictive value (precision), negative predictive value, accuracy and the area under the curve (AUC).

RESULTS

The area under curve (AUC) of support vector classifier (SVC), linear discrimination analysis (LDA), k-nearest neighbors (KNN), and gaussian naïve bias (NB) algorithms in training (testing) sets are 0.99 (0.94), 1.00 (0.87), 0.98 (0.92), 0.97 (0.97), and 0.98 (0.93), respectively. Nested cross-validation demonstrated the robustness and generalizability of the models. Additionally, the calibration plot and decision curve analysis demonstrated the practical usefulness of these models in clinical practice, with lower bias toward different subgroups during decision-making.

CONCLUSION

The combination of radiomics and machine learning approach based on contrast enhanced T1WI images could well distinguish CAE and BM. This approach holds promise in assisting doctors with accurate diagnosis and clinical decision-making.

Study on the predictive value of serum hypersensitive C-reactive protein, homocysteine, fibrinogen, and omentin-1 levels with gestational diabetes mellitus

Objective: To investigate whether hypersensitive C-reactive protein (Hs-CRP), homocysteine, fibrinogen, and omentin-1 could predict gestational diabetes mellitus (GDM) risk. Methods: Case-control study was conducted at Hengshui People's Hospital. The GDM group included data about 150 patients aged between 22 and 35 years in 24-28 weeks. An equivalent comparative control group without GDM was composed of the same pool of patients. Body mass index (BMI), total cholesterol (TC), triglyceride, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), oral glucose tolerance test (OGTT) 0-2h, hs-CRP, homocysteine, fibrinogen, and omentin-1 levels were studied in the serum samples of research groups. Univariate logistic regression analysis was used to explore the risk factors of GDM. The area under the curve (AUC) was calculated by the receiver operating characteristic curve (ROC) to analyze the predictive values. Results: Hs-CRP, homocysteine, and fibrinogen in GDM group were significantly higher than those in non-GDM group. Omentin-1 were significantly lower than those in non-GDM group. Logistic regression showed that hs-CRP, homocysteine, fibrinogen, and omentin-1 were risk factors for GDM. The AUC of the established GDM risk prediction model was 0.977, and the sensitivity and specificity were 92.10% and 98.70%, respectively; which were greater than that of hs-CRP, homocysteine, fibrinogen, and omentin-1 alone. Conclusions: Hs-CRP, homocysteine, fibrinogen, and omentin-1 in pregnancy have important clinical value for the prediction of GDM. We used these laboratory indications to establish a GDM risk prediction model that allows for early detection and treatment of GDM, lowering the morbidity of maternal and infant complications.

Outcomes of Infants Who Are Febrile Aged 29-90 Days Discharged from the Emergency Department

OBJECTIVE

To evaluate the characteristics and outcomes of infants aged 29-90 days who are febrile discharged from the pediatric emergency department (ED).

STUDY DESIGN

This was a multicenter, retrospective cohort study of infants aged 29-90 days who visited any of the 7 Clalit Health Services pediatric EDs in Israel between January 1, 2019, and March 31, 2022. Infants who were febrile discharged from the ED after having blood and urine cultures taken were included. The primary outcome measure was the incidence of return visit (RV) to an ED. Secondary outcome measures were the incidence of invasive bacterial infection, urinary tract infection, pediatric intensive care unit admissions, and deaths. We assessed variables associated with the primary outcomes.

RESULTS

A total of 1647 infants were included. Their median (IQR) age at ED visit was 58.5 (47.7, 72.7) days, 53.1% were male. A total of 329 patients (20%) returned to the ED within 120 hours. Overall, 7.8% of discharged infants had a positive urine culture, 4 (0.2%) had a positive blood culture, and none had meningitis. One patient was admitted to the pediatric intensive care unit, and there was no death. Abnormal C-reactive protein was associated with RV among 61- to 90-day-old infants.

CONCLUSIONS

Infants aged 29-90 days who were febrile and discharged following a protocol-driven pathway from the pediatric ED had a relatively high RV rate. However, the rate of urinary tract infection was relatively low, and rate of invasive bacterial infection was extremely low. There were no deaths or serious sequelae.

Prediction of gestational diabetes mellitus using machine learning from birth cohort data of the Japan Environment and Children's Study

Recently, prediction of gestational diabetes mellitus (GDM) using artificial intelligence (AI) from medical records has been reported. We aimed to evaluate GDM-predictive AI-based models using birth cohort data with a wide range of information and to explore factors contributing to GDM development. This investigation was conducted as a part of the Japan Environment and Children's Study. In total, 82,698 pregnant mothers who provided data on lifestyle, anthropometry, and socioeconomic status before pregnancy and the first trimester were included in the study. We employed machine learning methods as AI algorithms, such as random forest (RF), gradient boosting decision tree (GBDT), and support vector machine (SVM), along with logistic regression (LR) as a reference. GBDT displayed the highest accuracy, followed by LR, RF, and SVM. Exploratory analysis of the JECS data revealed that health-related quality of life in early pregnancy and maternal birthweight, which were rarely reported to be associated with GDM, were found along with variables that were reported to be associated with GDM. The results of decision tree-based algorithms, such as GBDT, have shown high accuracy, interpretability, and superiority for predicting GDM using birth cohort data.

Integration of clinical demographics and routine laboratory analysis parameters for early prediction of gestational diabetes mellitus in the Chinese population

Gestational diabetes mellitus (GDM) is one of the most common complications in pregnancy, impairing both maternal and fetal health in short and long term. As early interventions are considered desirable to prevent GDM, this study aims to develop a simple-to-use nomogram based on multiple common risk factors from electronic medical health records (EMHRs). A total of 924 pregnant women whose EMHRs were available at Peking University International Hospital from January 2022 to October 2022 were included. Clinical demographics and routine laboratory analysis parameters at 8-12 weeks of gestation were collected. A novel nomogram was established based on the outcomes of multivariate logistic regression. The nomogram demonstrated powerful discrimination (the area under the receiver operating characteristic curve = 0.7542), acceptable agreement (Hosmer-Lemeshow test, P = 0.3214) and favorable clinical utility. The C-statistics of 10-Fold cross validation, Leave one out cross validation and Bootstrap were 0.7411, 0.7357 and 0.7318, respectively, indicating the stability of the nomogram. A novel nomogram based on easily-accessible parameters was developed to predict GDM in early pregnancy, which may provide a paradigm for repurposing clinical data and benefit the clinical management of GDM. There is a need for prospective multi-center studies to validate the nomogram before employing the nomogram in real-world clinical practice.

Elevated levels of renal function tests conferred increased risks of developing various pregnancy complications and adverse perinatal outcomes: insights from a population-based cohort study

OBJECTIVES

Physiological changes during pregnancy can affect the results of renal function tests (RFTs). In this population-based cohort study, we aimed to establish trimester-specific reference intervals (RIs) of RFTs in singleton and twin pregnancies and systematically investigate the relationship between RFTs and adverse pregnancy outcomes.

METHODS

The laboratory results of the first- and third-trimester RFTs, including blood urea nitrogen (BUN), serum uric acid (UA), creatinine (Crea) and cystatin C (Cys C), and the relevant medical records, were retrieved from 29,328 singleton and 840 twin pregnant women who underwent antenatal examinations from November 20, 2017 to January 31, 2021. The trimester-specific RIs of RFTs were estimated with both of the direct observational and the indirect Hoffmann methods. The associations between RTFs and pregnancy complications as well as perinatal outcomes were assessed by logistic regression analysis.

RESULTS

Maternal RFTs showed no significant difference between the direct RIs established with healthy pregnant women and the calculated RIs derived from the Hoffmann method. In addition, elevated levels of RFTs were associated with increased risks of developing various pregnancy complications and adverse perinatal outcomes. Notably, elevated third-trimester RFTs posed strong risks of preterm birth (PTB) and fetal growth restriction (FGR).

CONCLUSIONS

We established the trimester-specific RIs of RFTs in both singleton and twin pregnancies. Our risk analysis findings underscored the importance of RFTs in identifying women at high risks of developing adverse complications or outcomes during pregnancy.

Early Prediction Model of Macrosomia Using Machine Learning for Clinical Decision Support

The condition of fetal overgrowth, also known as macrosomia, can cause serious health complications for both the mother and the infant. It is crucial to identify high-risk macrosomia-relevant pregnancies and intervene appropriately. Despite this need, there are several gaps in research related to macrosomia, including limited predictive models, insufficient machine learning applications, ineffective interventions, and inadequate understanding of how to integrate machine learning models into clinical decision-making. To address these gaps, we developed a machine learning-based model that uses maternal characteristics and medical history to predict macrosomia. Three different algorithms, namely logistic regression, support vector machine, and random forest, were used to develop the model. Based on the evaluation metrics, the logistic regression algorithm provided the best results among the three. The logistic regression algorithm was chosen as the final algorithm to predict macrosomia. The hyper parameters of the logistic regression model were tuned using cross-validation to achieve the best possible performance. Our results indicate that machine learning-based models have the potential to improve macrosomia prediction and enable appropriate interventions for high-risk pregnancies, leading to better health outcomes for both mother and fetus. By leveraging machine learning algorithms and addressing research gaps related to macrosomia, we can potentially reduce the health risks associated with this condition and make informed decisions about high-risk pregnancies.

Prediction of gestational diabetes mellitus in Asian women using machine learning algorithms

This study developed a machine learning algorithm to predict gestational diabetes mellitus (GDM) using retrospective data from 34,387 pregnancies in multi-centers of South Korea. Variables were collected at baseline, E0 (until 10 weeks' gestation), E1 (11-13 weeks' gestation) and M1 (14-24 weeks' gestation). The data set was randomly divided into training and test sets (7:3 ratio) to compare the performances of light gradient boosting machine (LGBM) and extreme gradient boosting (XGBoost) algorithms, with a full set of variables (original). A prediction model with the whole cohort achieved area under the receiver operating characteristics curve (AUC) and area under the precision-recall curve (AUPR) values of 0.711 and 0.246 at baseline, 0.720 and 0.256 at E0, 0.721 and 0.262 at E1, and 0.804 and 0.442 at M1, respectively. Then comparison of three models with different variable sets were performed: [a] variables from clinical guidelines; [b] selected variables from Shapley additive explanations (SHAP) values; and [c] Boruta algorithms. Based on model [c] with the least variables and similar or better performance than the other models, simple questionnaires were developed. The combined use of maternal factors and laboratory data could effectively predict individual risk of GDM using a machine learning model.

Predicting prenatal depression and assessing model bias using machine learning models

Perinatal depression (PND) is one of the most common medical complications during pregnancy and postpartum period, affecting 10-20% of pregnant individuals. Black and Latina women have higher rates of PND, yet they are less likely to be diagnosed and receive treatment. Machine learning (ML) models based on Electronic Medical Records (EMRs) have been effective in predicting postpartum depression in middle-class White women but have rarely included sufficient proportions of racial and ethnic minorities, which contributed to biases in ML models for minority women. Our goal is to determine whether ML models could serve to predict depression in early pregnancy in racial/ethnic minority women by leveraging EMR data. We extracted EMRs from a hospital in a large urban city that mostly served low-income Black and Hispanic women (N=5,875) in the U.S. Depressive symptom severity was assessed from a self-reported questionnaire, PHQ-9. We investigated multiple ML classifiers, used Shapley Additive Explanations (SHAP) for model interpretation, and determined model prediction bias with two metrics, Disparate Impact, and Equal Opportunity Difference. While ML model (Elastic Net) performance was low (ROCAUC=0.67), we identified well-known factors associated with PND, such as unplanned pregnancy and being single, as well as underexplored factors, such as self-report pain levels, lower levels of prenatal vitamin supplement intake, asthma, carrying a male fetus, and lower platelet levels blood. Our findings showed that despite being based on a sample mostly composed of 75% low-income minority women (54% Black and 27% Latina), the model performance was lower for these communities. In conclusion, ML models based on EMRs could moderately predict depression in early pregnancy, but their performance is biased against low-income minority women.

Desire to Avoid Pregnancy scale: clinical considerations and comparison with other questions about pregnancy preferences

BACKGROUND

Clinicians and women of reproductive age would benefit from a reliable way to identify who is likely to become pregnant in the next year, in order to direct health advice. The 14-item Desire to Avoid Pregnancy (DAP) scale is predictive of pregnancy; this paper compares it with other ways of assessing pregnancy preferences to shortlist options for clinical implementation.

METHODS

A cohort of 994 UK women of reproductive age completed the DAP and other questions about pregnancy preferences, including the Attitude towards Potential Pregnancy Scale (APPS), at baseline and reported on pregnancies quarterly for a year. For each question, DAP item and combinations of DAP items, we examined the predictive ability, sensitivity, specificity, area under the receiver operating curve (AUROC), and positive and negative predictive values.

RESULTS

The AUROCs and predictive ability of the APPS and DAP single items were weaker than the full DAP, though all except one had acceptable AUROCs (>0.7). The most predictive individual DAP item was 'It would be a good thing for me if I became pregnant in the next 3 months', where women who strongly agreed had a 66.7% chance of pregnancy within 12 months and the AUROC was acceptable (0.77).

CONCLUSION

We recommend exploring the acceptability to women and healthcare professionals of asking a single DAP item ('It would be a good thing for me if I became pregnant in the next 3 months'), possibly in combination with additional DAP items. This will help to guide service provision to support reproductive preferences.

Prediction of gestational diabetes using deep learning and Bayesian optimization and traditional machine learning techniques

The study aimed to develop a clinical diagnosis system to identify patients in the GD risk group and reduce unnecessary oral glucose tolerance test (OGTT) applications for pregnant women who are not in the GD risk group using deep learning algorithms. With this aim, a prospective study was designed and the data was taken from 489 patients between the years 2019 and 2021, and informed consent was obtained. The clinical decision support system for the diagnosis of GD was developed using the generated dataset with deep learning algorithms and Bayesian optimization. As a result, a novel successful decision support model was developed using RNN-LSTM with Bayesian optimization that gave 95% sensitivity and 99% specificity on the dataset for the diagnosis of patients in the GD risk group by obtaining 98% AUC (95% CI (0.95-1.00) and p < 0.001). Thus, with the clinical diagnosis system developed to assist physicians, it is planned to save both cost and time, and reduce possible adverse effects by preventing unnecessary OGTT for patients who are not in the GD risk group.

Development of machine learning models to predict gestational diabetes risk in the first half of pregnancy

BACKGROUND

Early prediction of Gestational Diabetes Mellitus (GDM) risk is of particular importance as it may enable more efficacious interventions and reduce cumulative injury to mother and fetus. The aim of this study is to develop machine learning (ML) models, for the early prediction of GDM using widely available variables, facilitating early intervention, and making possible to apply the prediction models in places where there is no access to more complex examinations.

METHODS

The dataset used in this study includes registries from 1,611 pregnancies. Twelve different ML models and their hyperparameters were optimized to achieve early and high prediction performance of GDM. A data augmentation method was used in training to improve prediction results. Three methods were used to select the most relevant variables for GDM prediction. After training, the models ranked with the highest Area under the Receiver Operating Characteristic Curve (AUCROC), were assessed on the validation set. Models with the best results were assessed in the test set as a measure of generalization performance.

RESULTS

Our method allows identifying many possible models for various levels of sensitivity and specificity. Four models achieved a high sensitivity of 0.82, a specificity in the range 0.72-0.74, accuracy between 0.73-0.75, and AUCROC of 0.81. These models required between 7 and 12 input variables. Another possible choice could be a model with sensitivity of 0.89 that requires just 5 variables reaching an accuracy of 0.65, a specificity of 0.62, and AUCROC of 0.82.

CONCLUSIONS

The principal findings of our study are: Early prediction of GDM within early stages of pregnancy using regular examinations/exams; the development and optimization of twelve different ML models and their hyperparameters to achieve the highest prediction performance; a novel data augmentation method is proposed to allow reaching excellent GDM prediction results with various models.

Generating synthetic mixed-type longitudinal electronic health records for artificial intelligent applications

The recent availability of electronic health records (EHRs) have provided enormous opportunities to develop artificial intelligence (AI) algorithms. However, patient privacy has become a major concern that limits data sharing across hospital settings and subsequently hinders the advances in AI. Synthetic data, which benefits from the development and proliferation of generative models, has served as a promising substitute for real patient EHR data. However, the current generative models are limited as they only generate single type of clinical data for a synthetic patient, i.e., either continuous-valued or discrete-valued. To mimic the nature of clinical decision-making which encompasses various data types/sources, in this study, we propose a generative adversarial network (GAN) entitled EHR-M-GAN that simultaneously synthesizes mixed-type timeseries EHR data. EHR-M-GAN is capable of capturing the multidimensional, heterogeneous, and correlated temporal dynamics in patient trajectories. We have validated EHR-M-GAN on three publicly-available intensive care unit databases with records from a total of 141,488 unique patients, and performed privacy risk evaluation of the proposed model. EHR-M-GAN has demonstrated its superiority over state-of-the-art benchmarks for synthesizing clinical timeseries with high fidelity, while addressing the limitations regarding data types and dimensionality in the current generative models. Notably, prediction models for outcomes of intensive care performed significantly better when training data was augmented with the addition of EHR-M-GAN-generated timeseries. EHR-M-GAN may have use in developing AI algorithms in resource-limited settings, lowering the barrier for data acquisition while preserving patient privacy.

Machine learning and disease prediction in obstetrics

Machine learning technologies and translation of artificial intelligence tools to enhance the patient experience are changing obstetric and maternity care. An increasing number of predictive tools have been developed with data sourced from electronic health records, diagnostic imaging and digital devices. In this review, we explore the latest tools of machine learning, the algorithms to establish prediction models and the challenges to assess fetal well-being, predict and diagnose obstetric diseases such as gestational diabetes, pre-eclampsia, preterm birth and fetal growth restriction. We discuss the rapid growth of machine learning approaches and intelligent tools for automated diagnostic imaging of fetal anomalies and to asses fetoplacental and cervix function using ultrasound and magnetic resonance imaging. In prenatal diagnosis, we discuss intelligent tools for magnetic resonance imaging sequencing of the fetus, placenta and cervix to reduce the risk of preterm birth. Finally, the use of machine learning to improve safety standards in intrapartum care and early detection of complications will be discussed. The demand for technologies to enhance diagnosis and treatment in obstetrics and maternity should improve frameworks for patient safety and enhance clinical practice.

Machine learning applied in maternal and fetal health: a narrative review focused on pregnancy diseases and complications

Introduction

Machine learning (ML) corresponds to a wide variety of methods that use mathematics, statistics and computational science to learn from multiple variables simultaneously. By means of pattern recognition, ML methods are able to find hidden correlations and accomplish accurate predictions regarding different conditions. ML has been successfully used to solve varied problems in different areas of science, such as psychology, economics, biology and chemistry. Therefore, we wondered how far it has penetrated into the field of obstetrics and gynecology.

Aim

To describe the state of art regarding the use of ML in the context of pregnancy diseases and complications.

Methodology

Publications were searched in PubMed, Web of Science and Google Scholar. Seven subjects of interest were considered: gestational diabetes mellitus, preeclampsia, perinatal death, spontaneous abortion, preterm birth, cesarean section, and fetal malformations.

Current state

ML has been widely applied in all the included subjects. Its uses are varied, the most common being the prediction of perinatal disorders. Other ML applications include (but are not restricted to) biomarker discovery, risk estimation, correlation assessment, pharmacological treatment prediction, drug screening, data acquisition and data extraction. Most of the reviewed articles were published in the last five years. The most employed ML methods in the field are non-linear. Except for logistic regression, linear methods are rarely used.

Future challenges

To improve data recording, storage and update in medical and research settings from different realities. To develop more accurate and understandable ML models using data from cutting-edge instruments. To carry out validation and impact analysis studies of currently existing high-accuracy ML models.

Conclusion

The use of ML in pregnancy diseases and complications is quite recent, and has increased over the last few years. The applications are varied and point not only to the diagnosis, but also to the management, treatment, and pathophysiological understanding of perinatal alterations. Facing the challenges that come with working with different types of data, the handling of increasingly large amounts of information, the development of emerging technologies, and the need of translational studies, it is expected that the use of ML continue growing in the field of obstetrics and gynecology.

Gestational diabetes is driven by microbiota-induced inflammation months before diagnosis

OBJECTIVE

Gestational diabetes mellitus (GDM) is a condition in which women without diabetes are diagnosed with glucose intolerance during pregnancy, typically in the second or third trimester. Early diagnosis, along with a better understanding of its pathophysiology during the first trimester of pregnancy, may be effective in reducing incidence and associated short-term and long-term morbidities.

DESIGN

We comprehensively profiled the gut microbiome, metabolome, inflammatory cytokines, nutrition and clinical records of 394 women during the first trimester of pregnancy, before GDM diagnosis. We then built a model that can predict GDM onset weeks before it is typically diagnosed. Further, we demonstrated the role of the microbiome in disease using faecal microbiota transplant (FMT) of first trimester samples from pregnant women across three unique cohorts.

RESULTS

We found elevated levels of proinflammatory cytokines in women who later developed GDM, decreased faecal short-chain fatty acids and altered microbiome. We next confirmed that differences in GDM-associated microbial composition during the first trimester drove inflammation and insulin resistance more than 10 weeks prior to GDM diagnosis using FMT experiments. Following these observations, we used a machine learning approach to predict GDM based on first trimester clinical, microbial and inflammatory markers with high accuracy.

CONCLUSION

GDM onset can be identified in the first trimester of pregnancy, earlier than currently accepted. Furthermore, the gut microbiome appears to play a role in inflammation-induced GDM pathogenesis, with interleukin-6 as a potential contributor to pathogenesis. Potential GDM markers, including microbiota, can serve as targets for early diagnostics and therapeutic intervention leading to prevention.

Development and validation of a screening questionnaire for early identification of pregnant women at risk for excessive gestational weight gain

BACKGROUND

Excessive weight gain during pregnancy is associated with adverse health outcomes for mother and child. Intervention strategies to prevent excessive gestational weight gain (GWG) should consider women's individual risk profile, however, no tool exists for identifying women at risk at an early stage. The aim of the present study was to develop and validate a screening questionnaire based on early risk factors for excessive GWG.

METHODS

The cohort from the German "Gesund leben in der Schwangerschaft"/ "healthy living in pregnancy" (GeliS) trial was used to derive a risk score predicting excessive GWG. Sociodemographics, anthropometrics, smoking behaviour and mental health status were collected before week 12^th of gestation. GWG was calculated using the last and the first weight measured during routine antenatal care. The data were randomly split into development and validation datasets with an 80:20 ratio. Using the development dataset, a multivariate logistic regression model with stepwise backward elimination was performed to identify salient risk factors associated with excessive GWG. The β coefficients of the variables were translated into a score. The risk score was validated by an internal cross-validation and externally with data from the FeLIPO study (GeliS pilot study). The area under the receiver operating characteristic curve (AUC ROC) was used to estimate the predictive power of the score.

RESULTS

1790 women were included in the analysis, of whom 45.6% showed excessive GWG. High pre-pregnancy body mass index, intermediate educational level, being born in a foreign country, primiparity, smoking, and signs of depressive disorder were associated with the risk of excessive GWG and included in the screening questionnaire. The developed score varied from 0-15 and divided the women´s risk for excessive GWG into low (0-5), moderate (6-10) and high (11-15). The cross-validation and the external validation yielded a moderate predictive power with an AUC of 0.709 and 0.738, respectively.

CONCLUSIONS

Our screening questionnaire is a simple and valid tool to identify pregnant women at risk for excessive GWG at an early stage. It could be used in routine care to provide targeted primary prevention measures to women at particular risk to gain excessive gestational weight.

TRIAL REGISTRATION

NCT01958307, ClinicalTrials.gov, retrospectively registered 9 October 2013.

A machine learning approach for early prediction of gestational diabetes mellitus using elemental contents in fingernails

The aim of this pilot study was to predict the risk of gestational diabetes mellitus (GDM) by the elemental content in fingernails and urine with machine learning analysis. Sixty seven pregnant women (34 control and 33 GDM patient) were included. Fingernails and urine were collected in the first and second trimesters, respectively. The concentrations of elements were determined by inductively coupled plasma-mass spectrometry. Logistic regression model was applied to estimate the adjusted odd ratios and 95% confidence intervals. The predictive performances of multiple machine learning algorithms were evaluated, and an ensemble model was built to predict the risk for GDM based on the elemental contents in the fingernails. Beryllium, selenium, tin and copper were positively associated with the risk of GDM while nickel and mercury showed opposite result. The trained ensemble model showed larger area under curve (AUC) of receiver operating characteristic curve (0.81) using fingernail Ni, Cu and Se concentrations. The model was validated by external data set with AUC = 0.71. In summary, the results of the present study highlight the potential of fingernails, as an alternative sample, together with machine learning in human biomonitoring studies.

DEVELOPMENT OF A MACHINE LEARNING MODEL FOR PREDICTING 28-DAY MORTALITY OF SEPTIC PATIENTS WITH ATRIAL FIBRILLATION

ABSTRACT

Introduction: Septic patients with atrial fibrillation (AF) are common in the intensive care unit accompanied by high mortality. The early prediction of prognosis of these patients is critical for clinical intervention. This study aimed to develop a model by using machine learning (ML) algorithms to predict the risk of 28-day mortality in septic patients with AF. Methods: In this retrospective cohort study, we extracted septic patients with AF from the Medical Information Mart for Intensive Care III (MIMIC-III) and IV database. Afterward, only MIMIC-IV cohort was randomly divided into training or internal validation set. External validation set was mainly extracted from MIMIC-III database. Propensity score matching was used to reduce the imbalance between the external validation and internal validation data sets. The predictive factors for 28-day mortality were determined by using multivariate logistic regression. Then, we constructed models by using ML algorithms. Multiple metrics were used for evaluation of performance of the models, including the area under the receiver operating characteristic curve, sensitivity, specificity, recall, and accuracy. Results: A total of 5,317 septic patients with AF were enrolled, with 3,845 in the training set, 960 in the internal testing set, and 512 in the external testing set, respectively. Then, we established four prediction models by using ML algorithms. AdaBoost showed moderate performance and had a higher accuracy than the other three models. Compared with other severity scores, the AdaBoost obtained more net benefit. Conclusion: We established the first ML model for predicting the 28-day mortality of septic patients with AF. Compared with conventional scoring systems, the AdaBoost model performed moderately. The model established will have the potential to improve the level of clinical practice.

Human-guided deep learning with ante-hoc explainability by convolutional network from non-image data for pregnancy prognostication

BACKGROUND AND OBJECTIVE

Deep learning is applied in medicine mostly due to its state-of-the-art performance for diagnostic imaging. Supervisory authorities also require the model to be explainable, but most explain the model after development (post hoc) instead of incorporating explanation into the design (ante hoc). This study aimed to demonstrate a human-guided deep learning with ante-hoc explainability by convolutional network from non-image data to develop, validate, and deploy a prognostic prediction model for PROM and an estimator of time of delivery using a nationwide health insurance database.

METHODS

To guide modeling, we constructed and verified association diagrams respectively from literatures and electronic health records. Non-image data were transformed into meaningful images utilizing predictor-to-predictor similarities, harnessing the power of convolutional neural network mostly used for diagnostic imaging. The network architecture was also inferred from the similarities.

RESULTS

This resulted the best model for prelabor rupture of membranes (n=883, 376) with the area under curves 0.73 (95% CI 0.72 to 0.75) and 0.70 (95% CI 0.69 to 0.71) respectively by internal and external validations, and outperformed previous models found by systematic review. It was explainable by knowledge-based diagrams and model representation.

CONCLUSIONS

This allows prognostication with actionable insights for preventive medicine.

Prediction of Postoperative Pulmonary Edema Risk Using Machine Learning

Postoperative pulmonary edema (PPE) is a well-known postoperative complication. We hypothesized that a machine learning model could predict PPE risk using pre- and intraoperative data, thereby improving postoperative management. This retrospective study analyzed the medical records of patients aged > 18 years who underwent surgery between January 2011 and November 2021 at five South Korean hospitals. Data from four hospitals (n = 221,908) were used as the training dataset, whereas data from the remaining hospital (n = 34,991) were used as the test dataset. The machine learning algorithms used were extreme gradient boosting, light-gradient boosting machine, multilayer perceptron, logistic regression, and balanced random forest (BRF). The prediction abilities of the machine learning models were assessed using the area under the receiver operating characteristic curve, feature importance, and average precisions of precision-recall curve, precision, recall, f1 score, and accuracy. PPE occurred in 3584 (1.6%) and 1896 (5.4%) patients in the training and test sets, respectively. The BRF model exhibited the best performance (area under the receiver operating characteristic curve: 0.91, 95% confidence interval: 0.84-0.98). However, its precision and f1 score metrics were not good. The five major features included arterial line monitoring, American Society of Anesthesiologists physical status, urine output, age, and Foley catheter status. Machine learning models (e.g., BRF) could predict PPE risk and improve clinical decision-making, thereby enhancing postoperative management.

Prediction of Suicidal Behaviors in the Middle-aged Population: Machine Learning Analyses of UK Biobank

BACKGROUND

Suicidal behaviors, including suicide deaths and attempts, are major public health concerns. However, previous suicide models required a huge amount of input features, resulting in limited applicability in clinical practice.

OBJECTIVE

We aimed to construct applicable models (ie, with limited features) for short- and long-term suicidal behavior prediction. We further validated these models among individuals with different genetic risks of suicide.

METHODS

Based on the prospective cohort of UK Biobank, we included 223 (0.06%) eligible cases of suicide attempts or deaths, according to hospital inpatient or death register data within 1 year from baseline and randomly selected 4460 (1.18%) controls (1:20) without such records. We similarly identified 833 (0.22%) cases of suicidal behaviors 1 to 6 years from baseline and 16,660 (4.42%) corresponding controls. Based on 143 input features, mainly including sociodemographic, environmental, and psychosocial factors; medical history; and polygenic risk scores (PRS) for suicidality, we applied a bagged balanced light gradient-boosting machine (LightGBM) with stratified 10-fold cross-validation and grid-search to construct the full prediction models for suicide attempts or deaths within 1 year or between 1 and 6 years. The Shapley Additive Explanations (SHAP) approach was used to quantify the importance of input features, and the top 20 features with the highest SHAP values were selected to train the applicable models. The external validity of the established models was assessed among 50,310 individuals who participated in UK Biobank repeated assessments both overall and by the level of PRS for suicidality.

RESULTS

Individuals with suicidal behaviors were on average 56 years old, with equal sex distribution. The application of these full models in the external validation data set demonstrated good model performance, with the area under the receiver operating characteristic (AUROC) curves of 0.919 and 0.892 within 1 year and between 1 and 6 years, respectively. Importantly, the applicable models with the top 20 most important features showed comparable external-validated performance (AUROC curves of 0.901 and 0.885) as the full models, based on which we found that individuals in the top quintile of predicted risk accounted for 91.7% (n=11) and 80.7% (n=25) of all suicidality cases within 1 year and during 1 to 6 years, respectively. We further obtained comparable prediction accuracy when applying these models to subpopulations with different genetic susceptibilities to suicidality. For example, for the 1-year risk prediction, the AUROC curves were 0.907 and 0.885 for the high (>2nd tertile of PRS) and low (<1st) genetic susceptibilities groups, respectively.

CONCLUSIONS

We established applicable machine learning-based models for predicting both the short- and long-term risk of suicidality with high accuracy across populations of varying genetic risk for suicide, highlighting a cost-effective method of identifying individuals with a high risk of suicidality.

Evaluation of first and second trimester maternal thyroid profile on the prediction of gestational diabetes mellitus and post load glycemia

Maternal thyroid alterations have been widely associated with the risk of gestational diabetes mellitus (GDM). This study aims to 1) test the first and the second trimester full maternal thyroid profile on the prediction of GDM, both alone and combined with non-thyroid data; and 2) make that prediction independent of the diagnostic criteria, by evaluating the effectiveness of the different maternal variables on the prediction of oral glucose tolerance test (OGTT) post load glycemia. Pregnant women were recruited in Concepción, Chile. GDM diagnosis was performed at 24-28 weeks of pregnancy by an OGTT (n = 54 for normal glucose tolerance, n = 12 for GDM). 75 maternal thyroid and non-thyroid parameters were recorded in the first and the second trimester of pregnancy. Various combinations of variables were assessed for GDM and post load glycemia prediction through different classification and regression machine learning techniques. The best predictive models were simplified by variable selection. Every model was subjected to leave-one-out cross-validation. Our results indicate that thyroid markers are useful for the prediction of GDM and post load glycemia, especially at the second trimester of pregnancy. Thus, they could be used as an alternative screening tool for GDM, independently of the diagnostic criteria used. The final classification models predict GDM with cross-validation areas under the receiver operating characteristic curve of 0.867 (p<0.001) and 0.920 (p<0.001) in the first and the second trimester of pregnancy, respectively. The final regression models predict post load glycemia with cross-validation Spearman r correlation coefficients of 0.259 (p = 0.036) and 0.457 (p<0.001) in the first and the second trimester of pregnancy, respectively. This investigation constitutes the first attempt to test the performance of the whole maternal thyroid profile on GDM and OGTT post load glycemia prediction. Future external validation studies are needed to confirm these findings in larger cohorts and different populations.

Artificial Intelligence for Predicting and Diagnosing Complications of Diabetes

Artificial intelligence can use real-world data to create models capable of making predictions and medical diagnosis for diabetes and its complications. The aim of this commentary article is to provide a general perspective and present recent advances on how artificial intelligence can be applied to improve the prediction and diagnosis of six significant complications of diabetes including (1) gestational diabetes, (2) hypoglycemia in the hospital, (3) diabetic retinopathy, (4) diabetic foot ulcers, (5) diabetic peripheral neuropathy, and (6) diabetic nephropathy.

Recent applications of machine learning and deep learning models in the prediction, diagnosis, and management of diabetes: a comprehensive review

Diabetes as a metabolic illness can be characterized by increased amounts of blood glucose. This abnormal increase can lead to critical detriment to the other organs such as the kidneys, eyes, heart, nerves, and blood vessels. Therefore, its prediction, prognosis, and management are essential to prevent harmful effects and also recommend more useful treatments. For these goals, machine learning algorithms have found considerable attention and have been developed successfully. This review surveys the recently proposed machine learning (ML) and deep learning (DL) models for the objectives mentioned earlier. The reported results disclose that the ML and DL algorithms are promising approaches for controlling blood glucose and diabetes. However, they should be improved and employed in large datasets to affirm their applicability.

Utilizing fog computing and explainable deep learning techniques for gestational diabetes prediction

Gestational diabetes mellitus (GDM) is one of the pregnancy complications that poses a significant risk on mothers and babies as well. GDM usually diagnosed at 22–26 of gestation. However, the early prediction is desirable as it may contribute to decrease the risk. The continuous monitoring for mother’s vital signs helps in predicting any deterioration during pregnancy. The originality of this paper is to provide comprehensive framework for pregnancy women monitoring. The proposed Data Replacement and Prediction Framework consists of three layers which are: (i) IoT Layer, (ii) Fog Layer, and (iii) Cloud Layer. The first layer used IOT sensors to aggregate vital sings from pregnancies using invasive and noninvasive sensors. Then the vital signs transmitted to fog nodes to processed and finally stored in the cloud layer. The main contribution in this paper is located in the fog layer producing GDM module to implement two influential tasks which are: (i) Data Finding Methodology (DFM), and (ii) Explainable Prediction Algorithm (EPM) using DNN. First, the DFM is used to replace the unused data to free the cache space for the new incoming data items. The cache replacement is very important in the case of healthcare system as the incoming vital signs are frequent and must be replaced continuously. Second, the EPM is used to predict the incidence of GDM that may occur in the second trimester of the pregnancy. To evaluate our model, we extract data of 16,354 pregnancy women from medical information mart for intensive care (MIMIC III) benchmark dataset. For each woman, vital signs, demographic data and laboratory tests was aggregated. The results of the prediction model superior the state of the art (ACC = 0.957, AUC = 0.942). Regarding to explainability, we utilized Shapley additive explanation framework to provide local and global explanation for the developed models. Overall, the proposed framework is medically intuitive, allow the early prediction of GDM with cost effective solution.

First-trimester glycosylated hemoglobin (HbA1c) and maternal characteristics in the prediction of gestational diabetes: An observational cohort study

INTRODUCTION

This study aimed to investigate the extent to which gestational diabetes mellitus (GDM) can be predicted in the first trimester by combining a marker of growing interest, glycosylated hemoglobin A1c (HbA1c), and maternal characteristics.

MATERIAL AND METHODS

This observational study was conducted in the outpatient obstetric department of our institution. The values of HbA1c and venous random plasma glucose were prospectively assessed in the first trimester of pregnancy. We determined maternal characteristics that were independent predictors from the regression analysis and calculated areas under the receiver-operating curves by combining the maternal age, body mass index, previous history of GDM, and first-degree family history for diabetes mellitus. Moreover we investigated the predictive capability of HbA1c to exclude GDM. Patients with a first-trimester HbA1c level of 6.5% (48 mmol/mol) or more were excluded. The study was registered at ClinicalTrials.gov ID: NCT02139254.

RESULTS

We included 785 cases with complete dataset. The prevalence of GDM was 14.7% (115/785). Those who developed GDM had significantly higher HbA1c and random plasma glucose values (p < 0.0001 and p = 0.0002, respectively). In addition, they had a higher body mass index, were more likely to have a history of GDM and/or a first-degree family history of diabetes. When these maternal characteristics were combined with the first-trimester HbA1c and random plasma glucose the combined area under the receiver operating characteristics curve was 0.76 (95% CI 0.70-0.81).

CONCLUSIONS

Our results indicate that HbA1c and random plasma glucose values combined with age, body mass index, and personal and family history, allow the identification of women in the first trimester who are at increased risk of developing GDM.

A Multifaceted benchmarking of synthetic electronic health record generation models

Synthetic health data have the potential to mitigate privacy concerns in supporting biomedical research and healthcare applications. Modern approaches for data generation continue to evolve and demonstrate remarkable potential. Yet there is a lack of a systematic assessment framework to benchmark methods as they emerge and determine which methods are most appropriate for which use cases. In this work, we introduce a systematic benchmarking framework to appraise key characteristics with respect to utility and privacy metrics. We apply the framework to evaluate synthetic data generation methods for electronic health records data from two large academic medical centers with respect to several use cases. The results illustrate that there is a utility-privacy tradeoff for sharing synthetic health data and further indicate that no method is unequivocally the best on all criteria in each use case, which makes it evident why synthetic data generation methods need to be assessed in context.

Risk prediction models of gestational diabetes mellitus before 16 gestational weeks

BACKGROUND

Gestational diabetes mellitus (GDM) can lead to adverse maternal and fetal outcomes, and early prevention is particularly important for their health, but there is no widely accepted approach to predict it in the early pregnancy. The aim of the present study is to build and evaluate predictive models for GDM using routine indexes, including maternal clinical characteristics and laboratory biomarkers, before 16 gestational weeks.

METHODS

A total of 2895 pregnant women were recruited and maternal clinical characteristics and laboratory biomarkers before 16 weeks of gestation were collected from two hospitals. All participants were randomly stratified into the training cohort and the internal validation cohort by the ratio of 7:3. Using multivariable logistic regression analysis, two nomogram models, including a basic model and an extended model, were built. The discrimination, calibration, and clinical validity were used to evaluate the models in the internal validation cohort.

RESULTS

The area under the receiver operating characteristic curve of the basic and the extended model was 0.736 and 0.756 in the training cohort, and was 0.736 and 0.763 in the validation cohort, respectively. The calibration curve analysis showed that the predicted values of the two models were not significantly different from the actual observations (p = 0.289 and 0.636 in the training cohort, p = 0.684 and 0.635 in the internal validation cohort, respectively). The decision-curve analysis showed a good clinical application value of the models.

CONCLUSIONS

The present study built simple and effective models, indicating that routine clinical and laboratory parameters can be used to predict the risk of GDM in the early pregnancy, and providing a novel reference for studying the prediction of GDM.

Development and Validation of Risk Prediction Models for Gestational Diabetes Mellitus Using Four Different Methods

Gestational diabetes mellitus (GDM), a common perinatal disease, is related to increased risks of maternal and neonatal adverse perinatal outcomes. We aimed to establish GDM risk prediction models that can be widely used in the first trimester using four different methods, including a score-scaled model derived from a meta-analysis using 42 studies, a logistic regression model, and two machine learning models (decision tree and random forest algorithms). The score-scaled model (seven variables) was established via a meta-analysis and a stratified cohort of 1075 Chinese pregnant women from the Northwest Women's and Children's Hospital (NWCH) and showed an area under the curve (AUC) of 0.772. The logistic regression model (seven variables) was established and validated using the above cohort and showed AUCs of 0.799 and 0.834 for the training and validation sets, respectively. Another two models were established using the decision tree (DT) and random forest (RF) algorithms and showed corresponding AUCs of 0.825 and 0.823 for the training set, and 0.816 and 0.827 for the validation set. The validation of the developed models suggested good performance in a cohort derived from another period. The score-scaled GDM prediction model, the logistic regression GDM prediction model, and the two machine learning GDM prediction models could be employed to identify pregnant women with a high risk of GDM using common clinical indicators, and interventions can be sought promptly.

Metabonomics fingerprint of volatile organic compounds in serum and urine of pregnant women with gestational diabetes mellitus

BACKGROUND

Gestational diabetes mellitus (GDM) is a metabolic disease with an increasing annual incidence rate. Our previous observational study found that pregnant women with GDM had mild cognitive decline.

AIM

To analyze the changes in metabonomics in pregnant women with GDM and explore the mechanism of cognitive function decline.

METHODS

Thirty GDM patients and 30 healthy pregnant women were analyzed. Solid-phase microextraction gas chromatography/mass spectrometry was used to detect organic matter in plasma and urine samples. Statistical analyses were conducted using principal component analysis and partial least squares discriminant analysis.

RESULTS

Differential volatile metabolites in the serum of pregnant women with GDM included hexanal, 2-octen-1-ol, and 2-propanol. Differential volatile metabolites in the urine of these women included benzene, cyclohexanone, 1-hexanol, and phenol. Among the differential metabolites, the conversion of 2-propanol to acetone may further produce methylglyoxal. Therefore, 2-propanol may be a potential marker for serum methylglyoxal.

CONCLUSION

2-propanol may be a potential volatile marker to evaluate cognitive impairment in pregnant women with GDM.

A machine learning model trained on a high-throughput antibacterial screen increases the hit rate of drug discovery

Screening for novel antibacterial compounds in small molecule libraries has a low success rate. We applied machine learning (ML)-based virtual screening for antibacterial activity and evaluated its predictive power by experimental validation. We first binarized 29,537 compounds according to their growth inhibitory activity (hit rate 0.87%) against the antibiotic-resistant bacterium Burkholderia cenocepacia and described their molecular features with a directed-message passing neural network (D-MPNN). Then, we used the data to train an ML model that achieved a receiver operating characteristic (ROC) score of 0.823 on the test set. Finally, we predicted antibacterial activity in virtual libraries corresponding to 1,614 compounds from the Food and Drug Administration (FDA)-approved list and 224,205 natural products. Hit rates of 26% and 12%, respectively, were obtained when we tested the top-ranked predicted compounds for growth inhibitory activity against B. cenocepacia, which represents at least a 14-fold increase from the previous hit rate. In addition, more than 51% of the predicted antibacterial natural compounds inhibited ESKAPE pathogens showing that predictions expand beyond the organism-specific dataset to a broad range of bacteria. Overall, the developed ML approach can be used for compound prioritization before screening, increasing the typical hit rate of drug discovery.

Predictive Modeling for the Diagnosis of Gestational Diabetes Mellitus Using Epidemiological Data in the United Arab Emirates

Gestational diabetes mellitus (GDM) is a common condition with repercussions for both the mother and her child. Machine learning (ML) modeling techniques were proposed to predict the risk of several medical outcomes. A systematic evaluation of the predictive capacity of maternal factors resulting in GDM in the UAE is warranted. Data on a total of 3858 women who gave birth and had information on their GDM status in a birth cohort were used to fit the GDM risk prediction model. Information used for the predictive modeling were from self-reported epidemiological data collected at early gestation. Three different ML models, random forest (RF), gradient boosting model (GBM), and extreme gradient boosting (XGBoost), were used to predict GDM. Furthermore, to provide local interpretation of each feature in GDM diagnosis, features were studied using Shapley additive explanations (SHAP). Results obtained using ML models show that XGBoost, which achieved an AUC of 0.77, performed better compared to RF and GBM. Individual feature importance using SHAP value and the XGBoost model show that previous GDM diagnosis, maternal age, body mass index, and gravidity play a vital role in GDM diagnosis. ML models using self-reported epidemiological data are useful and feasible in prediction models for GDM diagnosis amongst pregnant women. Such data should be periodically collected at early pregnancy for health professionals to intervene at earlier stages to prevent adverse outcomes in pregnancy and delivery. The XGBoost algorithm was the optimal model for identifying the features that predict GDM diagnosis.

Incidence, Characteristics, and Outcomes of Clinically Undetected Bacteremia in Children Discharged Home From the Emergency Department

BACKGROUND

Despite a recent decline in the rates of invasive infections, bacteremia in young children remains a significant challenge. We aimed to describe patient characteristics, microbial etiology, and outcomes of bacteremic, well-appearing children 3-36 months of age who were discharged home from the pediatric emergency department (PED) on their index visit.

METHODS

A retrospective cohort study in the PED of a tertiary children's hospital from 1 June 2015 until 30 June 2021. We included all well appearing, immunocompetent infants 3-36 months old evaluated for fever and discharged home from the PED after a blood culture was drawn. We extracted demographic, clinical and laboratory data from the patient's electronic medical records for the index visit and subsequent encounters.

RESULTS

During the study period, 17,114 children evaluated for fever met the inclusion criteria. Seventy-two patients (0.42%) had positive cultures for known pathogens. Thirty-six (50%) were male and 36 (50%) younger than 1 year. The most common isolates were S. pneumonia 26%. (n = 19), K. Kingae 25%. (n = 18) and Salmonella spp. 13.9% (n = 10). Sixty patients (85.7%) were recalled to the ED or had a scheduled appointment, 10 (14.3%) returned spontaneously and two were followed up by phone. The median time between visits was 28.7 hours (IQR 19.1-41.1). One patient was admitted to intensive care during the course of hospitalization. There were no deaths.

CONCLUSION

The rate of undetected true bacteremia in our study was low and our data suggest that significant clinical deterioration during the first 24 hours is rare.

Dense phenotyping from electronic health records enables machine learning-based prediction of preterm birth

BACKGROUND

Identifying pregnancies at risk for preterm birth, one of the leading causes of worldwide infant mortality, has the potential to improve prenatal care. However, we lack broadly applicable methods to accurately predict preterm birth risk. The dense longitudinal information present in electronic health records (EHRs) is enabling scalable and cost-efficient risk modeling of many diseases, but EHR resources have been largely untapped in the study of pregnancy.

METHODS

Here, we apply machine learning to diverse data from EHRs with 35,282 deliveries to predict singleton preterm birth.

RESULTS

We find that machine learning models based on billing codes alone can predict preterm birth risk at various gestational ages (e.g., ROC-AUC = 0.75, PR-AUC = 0.40 at 28 weeks of gestation) and outperform comparable models trained using known risk factors (e.g., ROC-AUC = 0.65, PR-AUC = 0.25 at 28 weeks). Examining the patterns learned by the model reveals it stratifies deliveries into interpretable groups, including high-risk preterm birth subtypes enriched for distinct comorbidities. Our machine learning approach also predicts preterm birth subtypes (spontaneous vs. indicated), mode of delivery, and recurrent preterm birth. Finally, we demonstrate the portability of our approach by showing that the prediction models maintain their accuracy on a large, independent cohort (5978 deliveries) from a different healthcare system.

CONCLUSIONS

By leveraging rich phenotypic and genetic features derived from EHRs, we suggest that machine learning algorithms have great potential to improve medical care during pregnancy. However, further work is needed before these models can be applied in clinical settings.

Development of early prediction model for pregnancy-associated hypertension with graph-based semi-supervised learning

Clinical guidelines recommend several risk factors to identify women in early pregnancy at high risk of developing pregnancy-associated hypertension. However, these variables result in low predictive accuracy. Here, we developed a prediction model for pregnancy-associated hypertension using graph-based semi-supervised learning. This is a secondary analysis of a prospective study of healthy pregnant women. To develop the prediction model, we compared the prediction performances across five machine learning methods (semi-supervised learning with both labeled and unlabeled data, semi-supervised learning with labeled data only, logistic regression, support vector machine, and random forest) using three different variable sets: [a] variables from clinical guidelines, [b] selected important variables from the feature selection, and [c] all routine variables. Additionally, the proposed prediction model was compared with placental growth factor, a predictive biomarker for pregnancy-associated hypertension. The study population consisted of 1404 women, including 1347 women with complete follow-up (labeled data) and 57 women with incomplete follow-up (unlabeled data). Among the 1347 with complete follow-up, 2.4% (33/1347) developed pregnancy-associated HTN. Graph-based semi-supervised learning using top 11 variables achieved the best average prediction performance (mean area under the curve (AUC) of 0.89 in training set and 0.81 in test set), with higher sensitivity (72.7% vs 45.5% in test set) and similar specificity (80.0% vs 80.5% in test set) compared to risk factors from clinical guidelines. In addition, our proposed model with graph-based SSL had a higher performance than that of placental growth factor for total study population (AUC, 0.71 vs. 0.80, p < 0.001). In conclusion, we could accurately predict the development pregnancy-associated hypertension in early pregnancy through the use of routine clinical variables with the help of graph-based SSL.

Association of Genetic Predisposition and Physical Activity With Risk of Gestational Diabetes in Nulliparous Women

IMPORTANCE

Polygenic risk scores (PRS) for type 2 diabetes (T2D) can improve risk prediction for gestational diabetes (GD), yet the strength of the association between genetic and lifestyle risk factors has not been quantified.

OBJECTIVE

To assess the association of PRS and physical activity in existing GD risk models and identify patient subgroups who may receive the most benefits from a PRS or physical activity intervention.

DESIGN, SETTINGS, AND PARTICIPANTS

The Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-Be cohort was established to study individuals without previous pregnancy lasting at least 20 weeks (nulliparous) and to elucidate factors associated with adverse pregnancy outcomes. A subcohort of 3533 participants with European ancestry was used for risk assessment and performance evaluation. Participants were enrolled from October 5, 2010, to December 3, 2013, and underwent genotyping between February 19, 2019, and February 28, 2020. Data were analyzed from September 15, 2020, to November 10, 2021.

EXPOSURES

Self-reported total physical activity in early pregnancy was quantified as metabolic equivalents of task (METs). Polygenic risk scores were calculated for T2D using contributions of 84 single nucleotide variants, weighted by their association in the Diabetes Genetics Replication and Meta-analysis Consortium data.

MAIN OUTCOMES AND MEASURES

Estimation of the development of GD from clinical, genetic, and environmental variables collected in early pregnancy, assessed using measures of model discrimination. Odds ratios and positive likelihood ratios were used to evaluate the association of PRS and physical activity with GD risk.

RESULTS

A total of 3533 women were included in this analysis (mean [SD] age, 28.6 [4.9] years). In high-risk population subgroups (body mass index ≥25 or aged ≥35 years), individuals with high PRS (top 25th percentile) or low activity levels (METs <450) had increased odds of a GD diagnosis of 25% to 75%. Compared with the general population, participants with both high PRS and low activity levels had higher odds of a GD diagnosis (odds ratio, 3.4 [95% CI, 2.3-5.3]), whereas participants with low PRS and high METs had significantly reduced risk of a GD diagnosis (odds ratio, 0.5 [95% CI, 0.3-0.9]; P = .01).

CONCLUSIONS AND RELEVANCE

In this cohort study, the addition of PRS was associated with the stratified risk of GD diagnosis among high-risk patient subgroups, suggesting the benefits of targeted PRS ascertainment to encourage early intervention.