Search

BackgroundMachine learning models require large datasets that may be siloed across different health care institutions. Machine learning studies that focus on COVID-19 have been limited to single-hospital data, which limits model generalizability. ObjectiveWe aimed to use federated learning, a machine learning technique that avoids locally aggregating raw clinical data across multiple institutions, to predict mortality in hospitalized patients with COVID-19 within 7 days. MethodsPatient data were collected from the electronic health records of 5 hospitals within the Mount Sinai Health System. Logistic regression with L1 regularization/least absolute shrinkage and selection operator (LASSO) and multilayer perceptron (MLP) models were trained by using local data at each site. We developed a pooled model with combined data from all 5 sites, and a federated model that only shared parameters with a central aggregator. ResultsThe LASSOfederated model outperformed the LASSOlocal model at 3 hospitals, and the MLPfederated model performed better than the MLPlocal model at all 5 hospitals, as determined by the area under the receiver operating characteristic curve. The LASSOpooled model outperformed the LASSOfederated model at all hospitals, and the MLPfederated model outperformed the MLPpooled model at 2 hospitals. ConclusionsThe federated learning of COVID-19 electronic health record data shows promise in developing robust predictive models without compromising patient privacy.

BackgroundCOVID-19 has infected millions of people worldwide and is responsible for several hundred thousand fatalities. The COVID-19 pandemic has necessitated thoughtful resource allocation and early identification of high-risk patients. However, effective methods to meet these needs are lacking. ObjectiveThe aims of this study were to analyze the electronic health records (EHRs) of patients who tested positive for COVID-19 and were admitted to hospitals in the Mount Sinai Health System in New York City; to develop machine learning models for making predictions about the hospital course of the patients over clinically meaningful time horizons based on patient characteristics at admission; and to assess the performance of these models at multiple hospitals and time points. MethodsWe used Extreme Gradient Boosting (XGBoost) and baseline comparator models to predict in-hospital mortality and critical events at time windows of 3, 5, 7, and 10 days from admission. Our study population included harmonized EHR data from five hospitals in New York City for 4098 COVID-19–positive patients admitted from March 15 to May 22, 2020. The models were first trained on patients from a single hospital (n=1514) before or on May 1, externally validated on patients from four other hospitals (n=2201) before or on May 1, and prospectively validated on all patients after May 1 (n=383). Finally, we established model interpretability to identify and rank variables that drive model predictions. ResultsUpon cross-validation, the XGBoost classifier outperformed baseline models, with an area under the receiver operating characteristic curve (AUC-ROC) for mortality of 0.89 at 3 days, 0.85 at 5 and 7 days, and 0.84 at 10 days. XGBoost also performed well for critical event prediction, with an AUC-ROC of 0.80 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. In external validation, XGBoost achieved an AUC-ROC of 0.88 at 3 days, 0.86 at 5 days, 0.86 at 7 days, and 0.84 at 10 days for mortality prediction. Similarly, the unimputed XGBoost model achieved an AUC-ROC of 0.78 at 3 days, 0.79 at 5 days, 0.80 at 7 days, and 0.81 at 10 days. Trends in performance on prospective validation sets were similar. At 7 days, acute kidney injury on admission, elevated LDH, tachypnea, and hyperglycemia were the strongest drivers of critical event prediction, while higher age, anion gap, and C-reactive protein were the strongest drivers of mortality prediction. ConclusionsWe externally and prospectively trained and validated machine learning models for mortality and critical events for patients with COVID-19 at different time horizons. These models identified at-risk patients and uncovered underlying relationships that predicted outcomes.

Some individuals with autism spectrum disorder (ASD) carry functional mutations rarely observed in the general population. We explored the genes disrupted by these variants from joint analysis of protein-truncating variants (PTVs), missense variants and copy number variants (CNVs) in a cohort of 63,237 individuals. We discovered 72 genes associated with ASD at false discovery rate (FDR) ≤ 0.001 (185 at FDR ≤ 0.05). De novo PTVs, damaging missense variants and CNVs represented 57.5%, 21.1% and 8.44% of association evidence, while CNVs conferred greatest relative risk. Meta-analysis with cohorts ascertained for developmental delay (DD) (n = 91,605) yielded 373 genes associated with ASD/DD at FDR ≤ 0.001 (664 at FDR ≤ 0.05), some of which differed in relative frequency of mutation between ASD and DD cohorts. The DD-associated genes were enriched in transcriptomes of progenitor and immature neuronal cells, whereas genes showing stronger evidence in ASD were more enriched in maturing neurons and overlapped with schizophrenia-associated genes, emphasizing that these neuropsychiatric disorders may share common pathways to risk.

The tensile bond strength between zirconia subjected to different surface-pretreatment methods and methacryloyloxydecyl-dihydrogen-phosphate (MDP)-containing self-adhesive resin cement was evaluated herein. Eighty-eight cylindrical zirconia specimens were randomly divided into the following four groups based on the pretreatment method: (1) no treatment, (2) air abrasion, (3) HNO3/HF etching, and (4) zirconia-nanoparticle coating. The tensile bond strength of the zirconia−resin-cement complexes was investigated. One-way ANOVA and post hoc tests were performed at a 95% significance level, and the Weibull modulus was calculated. Fracture patterns were visualized by SEM. The surface roughness of the specimens without resin bonding was evaluated by AFM. The tensile bond strength of the specimens decreased as follows: Groups 3 > 4 > 2 > 1 (28.2 ± 6.6, 26.1 ± 5.7, 16.6 ± 3.3, and 13.9 ± 3.0 MPa, respectively). Groups 3 and 4 had significantly higher tensile bond strengths (p < 0.05) and lower fracture probabilities than those of Groups 1 and 2. They also showed both mixed failure and resin-cement cohesive failure, whereas Groups 1 and 2 showed mixed failure exclusively. The zirconia−resin tensile bond was stronger after HNO3/HF etching or ZrO2-nanoparticle coating than after air abrasion or no treatment. The estimated surface roughness decreased as follows: Groups 3 > 4 > 2 > 1. The combination of zirconia pretreated with HNO3/HF etching or ZrO2-nanoparticle coating and an MDP-containing self-adhesive resin cement can increase the clinical longevity of zirconia restorations by preventing their decementation.

Providing assessment validity information to decision makers in a clear and useful format is an ongoing challenge for the educational and psychological measurement community. We identify issues with a previous approach to a graphical presentation, noting that it is mislabeled as presenting incremental validity, when in fact it displays the effects of using predictors in a multiple hurdle fashion. We offer a straightforward technique for displaying incremental validity among predictors in reference to a criterion measure.

BackgroundExcessive gestational weight gain (EGWG) closely associates with postpartum obesity. However, the causal role of EGWG in postpartum obesity has not been experimentally verified. The objective of this study was to determine whether and how EGWG causes long-term postpartum obesity.MethodsC57BL/6 mice were fed with high-fat diet during gestation (HFFDG) or control chow, then their body composition and energy metabolism were monitored after delivery.ResultsWe found that HFFDG significantly increased gestational weight gain. After delivery, adiposity of HFFDG-treated mice (Preg-HF) quickly recovered to the levels of controls. However, 3 months after parturition, Preg-HF mice started to gain significantly more body fat even with regular chow. The increase of body fat of Preg-HF mice was progressive with aging and by 9 months after delivery had increased 2-fold above the levels of controls. The expansion of white adipose tissue (WAT) of Preg-HF mice was manifested by hyperplasia in visceral fat and hypertrophy in subcutaneous fat. Preg-HF mice developed low energy expenditure and UCP1 expression in interscapular brown adipose tissue (iBAT) in later life. Although blood estrogen concentrations were similar between Preg-HF and control mice, a significant decrease in estrogen receptor α (ERα) expression and hypermethylation of the ERα promoter was detected in the fat of Preg-HF mice 9 months after delivery. Interestingly, hypermethylation of ERα promoter and low ERα expression were only detected in adipocyte progenitor cells in both iBAT and WAT of Preg-HF mice at the end of gestation.ConclusionsThese results demonstrate that HFFDG causes long-term postpartum obesity independent of early postpartum fat retention. This study also suggests that HFFDG adversely programs long-term postpartum energy metabolism by epigenetically reducing estrogen signaling in both BAT and WAT.

Autism spectrum disorder (ASD) is a markedly heterogeneous condition with a varied phenotypic presentation. Its high concordance among siblings, as well as its clear association with specific genetic disorders, both point to a strong genetic etiology. However, the molecular basis of ASD is still poorly understood, although recent studies point to the existence of sex-specific ASD pathophysiologies and biomarkers. Despite this, little is known about how exactly sex influences the gene expression signatures of ASD probands. In an effort to identify sex-dependent biomarkers and characterize their function, we present an analysis of a single paired-end postmortem brain RNA-Seq data set and a meta-analysis of six blood-based microarray data sets. Here, we identify several genes with sex-dependent dysregulation, and many more with sex-independent dysregulation. Moreover, through pathway analysis, we find that these sex-independent biomarkers have substantially different biological roles than the sex-dependent biomarkers, and that some of these pathways are ubiquitously dysregulated in both postmortem brain and blood. We conclude by synthesizing the discovered biomarker profiles with the extant literature, by highlighting the advantage of studying sex-specific dysregulation directly, and by making a call for new transcriptomic data that comprise large female cohorts.

BackgroundNumerous pathologies of pregnancy originate from placental dysfunction. It is essential to understand the functions of key genes in the placenta in order to discern the etiology of placental pathologies. A paucity of animal models that allow conditional and inducible expression of a target gene in the placenta is a major limitation for studying placental development and function.ResultsTo study the platelet-derived growth factor receptor alpha (PDGFRα)-directed and tamoxifen-induced Cre recombinase expression in the placenta, PDGFRα-CreER mice were crossed with mT/mG dual-fluorescent reporter mice. The expression of endogenous membrane-localized enhanced green fluorescent protein (mEGFP) and/or dTomato in the placenta was examined to identify PDGFRα promoter-directed Cre expression. Pregnant PDGFRα-CreER;mT/mG mice were treated with tamoxifen at various gestational ages. Upon tamoxifen treatment, reporter protein mEGFP was observed in the junctional zone (JZ) and chorionic plate (CP). Furthermore, a single dose of tamoxifen was sufficient to induce the recombination.ConclusionsPDGFRα-CreER expression is restricted to the JZ and CP of mouse placentas. PDGFRα-CreER mice provide a useful tool to conditionally knock out or overexpress a target gene in these regions of the mouse placenta.

Students bring with them rich cultural and linguistic ways of knowing and communicating when engaged in figuring out explanations of phenomena. However, it is a challenge for teachers to support bi/multilingual learners in language-intensive science and engineering practices when they don't share a common language. When we position students' bi/multilingualism as an asset rather than an obstacle, we can support bi/multilingual students' equitable sensemaking. To support bi/multilingual learners' equitable sensemaking, we offer two tools: language surveys and the multiple ways of communicating strategies sheet. We describe how a teacher (second author) used these resources to adapt an investigation about air resistance and contact forces for his bi/multilingual students. The tools and examples provided can help educators better understand their students and make principled changes to science instruction, to expand how we use language in science. We hope the tools aid teachers in this challenging work but also reframe how educators see what students are saying and doing as assets for developing their ideas and learning. [ABSTRACT FROM AUTHOR]

Binocular diplopia limits quality of life by hampering visual function. There are several treatment options with individual advantages and drawbacks. Contact lenses have recently emerged as an attractive treatment delivery mechanism due to advances in lens manufacturing and imaging modalities. Contact lenses can offer better comfort, efficacy, and cosmetic appeal compared to alternative treatments such as spectacles. They also present unique challenges due to risks of inflammation, infection, and intolerability. This review will explore the management of binocular diplopia using several types of contact lenses including base down prism in contact lenses, scleral lenses, occlusive lenses, and monovision lenses. Increased awareness of differing contact lens will help providers and patients make better informed decisions about diplopia management. [ABSTRACT FROM AUTHOR]

This study describes decentralized recruitment and enrollment for a COVID-19 treatment trial, while comparing 5 primary recruitment methods: search engine ads, paid advertising within a national testing company, paid advertising within a regional testing company, electronic health record messages, and word of mouth. These are compared across patient demographics, efficiency, and cost. Clinical Trials Registration NCT04510194. [ABSTRACT FROM AUTHOR]