Sebastian Enrique Illanes

Background: Prediction of gestational diabetes mellitus (GDM) in early pregnancy enables timely, targeted interventions to reduce risks for both the mother and fetus. We aimed to improve early GDM prediction and clinical risk stratification using a novel deep learning (DL) approach that transforms tabular clinical data into images using a genetic algorithm. Methods: A dataset of 1,611 pregnancies was used. The dataset contained routinely collected early-pregnancy data, including clinical measurements, blood tests, and medical history. A novel method to transform tabular data into images was applied to enable the use of deep learning models for GDM prediction. A genetic algorithm optimized the spatial arrangement of variables during image generation by determining their spatial positions. Interpretability was assessed using gradient-weighted class activation mapping (Grad-CAM) to identify variables the model preferentially attended to for positive and negative predictions. Results: Multiple model configurations were evaluated by varying the spatial arrangement of variables within the transformation. Prioritizing high sensitivity, one model achieved an accuracy of 0.86, sensitivity of 0.82, specificity of 0.87, and an AUC-ROC of 0.83. Another model achieved an accuracy of 0.79, sensitivity of 0.92, specificity of 0.77, and an AUC-ROC of 0.85. Compared with previously published machine learning models using the same dataset, our approach demonstrated improved performance. Our best models increased specificity by an average of 13.93 percentage points compared with traditional ML models at matched sensitivity levels, thereby reducing false-positive classifications that may lead to additional clinical evaluation. Grad-CAM analysis revealed that attention patterns of the model aligned with established clinical risk factors for GDM, with first trimester fasting glycemia and parity preferentially attended for positive predictions and maternal weight for negative predictions. Conclusions: This study demonstrates that transforming tabular clinical data into image representations, optimized using a genetic algorithm, improves early GDM prediction. Representing tabular variables as images enables more effective application of DL models, which have demonstrated enhanced pattern recognition in image-based tasks compared with traditional machine learning methods. Performance improvements over state-of-the-art approaches were achieved. Accurate DL models can be developed using a limited number of input variables, supporting early identification of women at increased risk of GDM and facilitating timely clinical management.

Background: Maternal outcomes remain inequitable worldwide. Severe morbidity persists, and current risk assessment tools are largely arbitrary, focusing on biomedical factors while overlooking social determinants of health. There is a need for data-driven AI models to improve early pregnancy risk identification and management. Objective: The study aimed to develop and internally validate first-trimester AI-based antenatal risk assessment models across three geographically and socioethnically diverse populations (Sweden, Chile, and Singapore) and to compare their performance with existing clinical risk assessment strategies. Methods: We conducted a retrospective population-based study using routinely collected first-trimester data from over 700,000 pregnancies from Sweden, Chile, and Singapore. Separate machine learning models predicting a composite of adverse maternal and neonatal outcomes were trained and internally validated for each population. Input variables were limited to information available at or before 14 weeks’ gestation. Model discrimination, measured by the area under the receiver operating characteristic (AUROC) curve, was compared with corresponding proxies for real-world first-trimester risk assessment approaches in each setting. Model interpretability was assessed using Shapley additive explanations. Results: The prevalence of the composite adverse outcome was 10.40% (75,647/727,354) in Sweden, 21.94% (1302/5934) in Chile, and 16.25% (6145/37,813) in Singapore. In Sweden, the guideline-based risk assessment achieved an AUROC of 0.53, compared with 0.65 for the LightGBM (light gradient boosting machine) model ( P <.001). In Chile, the midwifery-led risk assessment achieved an AUROC of 0.52, versus 0.65 from the traditional machine learning LightGBM model ( P <.001). In Singapore, the health care professional–based risk assessment reached an AUROC of 0.56, compared with 0.60 for the LightGBM model ( P <.05). In the Swedish and Singapore cohorts, sociodemographic variables were among the most influential predictive features. At a false-positive rate of 50.0%, the sensitivities for predicting the primary composite adverse outcome were 69.17%, 69.62%, and 64.93% for the Sweden, Chile, and Singapore models, respectively. At a 30.0% false-positive rate operating point, the sensitivities were lower, at 50.38%, 55.38%, and 44.75%, respectively, but with higher positive predictive values of 16.31%, 34.53%, and 22.32%, respectively. The model calibration plots showed reasonable agreement in Sweden and Singapore, whereas the Chile model showed poorer calibration, with a calibration slope of approximately 1.45, indicating underconfidence. Conclusions: AI-based models developed using first-trimester data generally demonstrated improved performance compared with existing first-trimester clinical risk stratification strategies across three distinct populations. These findings suggest the potential feasibility of population-specific, AI-enabled risk stratification as a clinical decision support tool and highlight the potential value of integrating social, demographic, and behavioral determinants into antenatal risk assessment frameworks to support more equitable and personalized antenatal care.

Prediction of Large for Gestational Age (LGA) risk is important as it can enable earlier, more effective interventions, and avoid or mitigate cumulative injury to both mother and baby at the time of delivery. The goal of this research is to improve the prediction of LGA using machine learning (ML) models and variables that are widely available as that allow for broad intervention in late pregnancy and during delivery. An improved prediction of LGA was achieved using twelve ML models with hyperparameter optimization. Also, to improve the LGA prediction a data augmentation method was employed in the training set. Additionally, improvement in LGA prediction was obtained with four variable selection methods employed to identify the most significant variables. To rank the best models on the validation set after training, the Area under the Receiver Operating Characteristic Curve (AUROC) was used. Finally, to assess the generalization performance, the best models were evaluated on the test set. Our method enabled us to identify several models with high sensitivity and specificity. The best models included those that achieved a sensitivity of 0.84, a specificity of 0.84, an accuracy of 0.84, and AUCROC 0.83, requiring 14 variables. Another model reached an accuracy of 0.87, a sensitivity of 0.71, and a specificity of 0.90 with an AUCROC of 0.83 (14 variables). Additionally, a model with a sensitivity of 0.58, the same as that described by Hadlock et al. [1] for ultrasound, required 10 variables, and reached an accuracy of 0.91, a specificity of 0.97, and an AUCROC of 0.86. Both models included maternal BMI (body mass index), First Control, Maternal Weight, BMI Last Control, and EFW (estimated fetal weight) as the most important variables. The main contributions of our study include the prediction of LGA using data obtained from standard clinical evaluations during prenatal care, as well as the development ML models, tuning the hyperparameters to improve prediction results, thus achieving high levels of sensitivity and specificity. To achieve optimal LGA prediction outcomes across different models, a data augmentation approach is introduced, an important improvement in LGA prediction over using only Hadlock's ultrasound formula.

The establishment of a successful pregnancy depends on proper embryo development and coordinated endometrial differentiation, particularly through the process of decidualization. Mesenchymal stem/stromal cells (MSCs), especially those derived from endometrial tissue (eMSCs) and menstrual fluid (MenSCs), have emerged as critical players in regulating decidualization, trophoblast invasion, angiogenesis, and immune modulation at the maternal-fetal interface. These cells exhibit potent immunoregulatory functions, underscoring their role in establishing a tolerant yet responsive environment for implantation along with promoting angiogenesis and stromal regeneration. Dysfunctional MSCs have been linked to impaired decidualization and pregnancy disorders, including preeclampsia (PE) and fetal growth restriction. Evidence indicates that MenSCs from women with a history of PE exhibit diminished angiogenic potential, impaired trophoblast invasion support, and altered cytokine secretion, associating these alterations to disease pathogenesis. Similarly, aberrant eMSCs contribute to endometriosis by supporting ectopic implantation, impaired decidualization, and enhanced migration.Collectively, these findings highlight the central role of endometrial- and menstrual-derived MSCs in both reproductive success and pathology. They also emphasize their potential as diagnostic biomarkers and therapeutic targets in obstetrics and gynecology. In this review, we summarize current knowledge on the role of endometrial- and menstrual-derived MSCs in pregnancy establishment and related complications, and discuss their emerging promise as diagnostic tools and therapeutic strategies in reproductive medicine.

During human pregnancy, plasma total cholesterol (TC) levels increase as pregnancy progresses, a phenomenon referred to as maternal physiological hyperchol

This study identifies a coexpression network module connecting birth weight with cord blood indole metabolites, highlighting an integrated neonatal growth-metabolism signature. Maternal lipid status tracks with this signature, adding molecular resolution to the developmental origins of health and disease (DOHaD) paradigm by implicating growth-linked biochemical perturbations in gut-derived indole metabolites. Given the established relevance of indole metabolites to cardiovascular health, our preliminary results motivate future studies aimed at targeting neonatal metabolism as an early window for cardiovascular health intervention.

Dyslipidemia, characterized by an excessive amount of lipids in the bloodstream, is a significant risk factor for metabolic disorders and cardiovascular diseases (CVDs). Maternal supraphysiological hypercholesterolemia (MSPH) is associated with increased maternal levels of total cholesterol (TC) and low‐density lipoprotein (LDL). This condition has been linked to negative consequences on the fetoplacental vasculature, including increased atherosclerosis development in the fetal aorta and later in children and adolescents. This study aims to determine whether the high cholesterol levels associated with MSPH affect lysosomal and mitochondrial functions in the syncytiotrophoblast (STB), considering the increased free cholesterol levels previously reported in primary human trophoblast (PHT) cells from MSPH pregnancies. Total cholesterol levels were measured in placental tissues and BeWo cells. Lysosomal mass, size, and activity, as well as mitochondrial mass, function, and morphology, were assessed in BeWo cells and placentas. Our results revealed that placental tissues from MSPH pregnancies and BeWo cells treated with oxidized (ox‐LDL) exhibited increased free cholesterol levels and higher expression of cholesterol transport proteins. Treatment of BeWo cells with ox‐LDL also led to an increase in lysosomal mass and size, accompanied by a decrease in lysosomal activity. Conversely, ox‐LDL treatment induced mitochondrial fragmentation in BeWo cells, together with reduced ATP production and diminished mitochondrial membrane potential. Similar alterations in lysosomes and mitochondria were observed in the placenta of patients with a history of MSPH. MSPH‐related high cholesterol levels induced by ox‐LDL impair lysosomal and mitochondrial functions in the STB, potentially contributing to cellular dysfunction observed in MSPH. This study highlights the importance of understanding the underlying mechanisms of MSPH to improve maternal and fetal health outcomes.

Background: Epidemiological studies suggest a link between gestational diabetes mellitus (GDM) and periodontitis. In this context, Porphyromonas gingivalis ( Pg ) and its byproducts, including lipopolysaccharide ( Pg ‐LPS), may translocate to the placenta, potentially intensifying pro‐inflammatory mechanisms in a hyperglycemic environment. This study aimed to assess the pro‐inflammatory simultaneous stimulus effect of hyperglycemia (HG) and Pg ‐LPS in human term placental explants. Methods: Fresh placental explants from healthy ( n = 10) and GDM ( n = 5) pregnant women were stimulated under normoglycemia (NG), hyperglycemia (HG), Pg ‐LPS, and a dual stimulus of HG+ Pg ‐LPS. Both placental explants and culture supernatants were assessed using histology, real‐time quantitative polymerase chain reaction (RT‐qPCR), Western blot, immunofluorescence, and multiplex immunoassay to analyze histopathological alterations, Toll‐like receptor (TLR)−2, −4, −9 mRNA expression, nuclear factor kappa‐light‐chain‐enhancer of activated B cells (NF‐κB) phosphorylation and nuclear immunolocalization, NLRP3 inflammasome expression, and pro‐inflammatory cytokines (interleukin [IL]‐6, IL‐1β, and tumor necrosis factor [TNF]‐α). Results: The dual stimulus (HG+ Pg ‐LPS) increased fibrinoid necrosis ( p  < 0.05) and upregulated TLR‐4 mRNA ( p  < 0.05) expression in placental explants, compared to other experimental conditions. NF‐κB phosphorylation and nuclear immunolocalization, NLRP3 inflammasome and IL‐6, IL‐1β, and TNF‐α mRNA expression increased upon the dual stimulus in healthy and GDM explants ( p  < 0.05). Additionally, levels of IL‐6 ( p  < 0.01), IL‐1β ( p  < 0.01), and TNF‐α ( p  < 0.0001) increased in the supernatant from healthy and GDM explants under the dual stimulus. No differences in cytokine levels were observed between healthy and GDM placental explants under the dual stimulus. Conclusions: A hyperglycemic environment amplifies the pro‐inflammatory response to Pg ‐LPS in human placental explants, suggesting that hyperglycemia synergizes with Pg ‐LPS in the placenta. Plain language summary: Bacteria that drive periodontitis, or their products, can enter the bloodstream and reach the placenta. In a high‐glucose environment, this exposure can intensify placental inflammation. This study aimed to determine whether exposure to a high‐glucose milieu, combined with byproducts of periodontal bacteria, activates inflammatory pathways in the placenta. Our results demonstrate that a high‐glucose environment amplifies the inflammatory response to periodontal bacteria in human placentas from both healthy pregnancies and those complicated by gestational diabetes mellitus. These findings advance our understanding of the connection between periodontitis and gestational diabetes mellitus and underscore the role of oral bacterial translocation in placental tissues during pregnancy.

Objective: To study variation in moral attitudes toward selective abortion in cases of prenatal diagnosis of a disability in the general population. Methods: 926 participants completed a survey in which they expressed their moral views regarding a person's decision to have an abortion following a prenatal diagnosis of a disability. Results: people morally endorsed selective abortion over the control condition (no diagnosis of disability); diagnoses of motor disability and sensorial disabilities were associated with stronger support for selective abortion; variations in prognoses about the expected social relationality and physical pain did not alter substantially the moral acceptance of selective abortion. Conclusions: people view selective abortion after a prenatal diagnosis of a disability as more morally acceptable than abortion without a diagnosis; however, predicting future impairments does not appear to influence support for pregnancy termination.

This study identifies a transcriptomic profile of long noncoding RNAs in gingival crevicular fluid samples in pregnant women with gestational diabetes risk. NEAT1 and LINC-PINT were increased expression in gingival crevicular fluid samples in pregnancies later diagnosed with gestational diabetes mellitus.