Search

We present an ensemble transfer learning method to predict suicide from Veterans Affairs (VA) electronic medical records (EMR). A diverse set of base models was trained to predict a binary outcome constructed from reported suicide, suicide attempt, and overdose diagnoses with varying choices of study design and prediction methodology. Each model used twenty cross-sectional and 190 longitudinal variables observed in eight time intervals covering 7.5 years prior to the time of prediction. Ensembles of seven base models were created and fine-tuned with ten variables expected to change with study design and outcome definition in order to predict suicide and combined outcome in a prospective cohort. The ensemble models achieved c-statistics of 0.73 on 2-year suicide risk and 0.83 on the combined outcome when predicting on a prospective cohort of [Formula: see text] 4.2 M veterans. The ensembles rely on nonlinear base models trained using a matched retrospective nested case-control (Rcc) study cohort and show good calibration across a diversity of subgroups, including risk strata, age, sex, race, and level of healthcare utilization. In addition, a linear Rcc base model provided a rich set of biological predictors, including indicators of suicide, substance use disorder, mental health diagnoses and treatments, hypoxia and vascular damage, and demographics.

Abstract Eucalyptus species play an important role in the global carbon cycle, especially in reducing the greenhouse effect as well as storing atmospheric CO₂. Thus, assessing the amount of CO₂ released by the soil in forest areas can generate important information for environmental monitoring. This study aims to verify the relation between soil carbon dioxide (CO₂) flux (FCO₂), spectral bands, and vegetation indices (VIs) derived from a UAV-based multispectral camera over an area of eucalyptus species. Multispectral imageries (green, red-edge, and near-infrared) from the Parrot Sequoia sensor, derived vegetation indices, and the FCO₂ data from a LI-COR 8100 analyzer, combined with soil moisture and temperature data, were collected and related. The vegetation indices ATSAVI (Adjusted Transformed Soil-Adjusted VI), GSAVI (Green Soil Adjusted Vegetation Index), and SAVI (Soil-Adjusted Vegetation Index), which use soil correction factors, exhibited a strong negative correlation with FCO₂ for the species E. camaldulensis, E. saligna, and E. urophylla species. A Multivariate Analysis of Variance showed significance (p

Abstract Scaling relationships are key in characterizing complex systems at criticality. In the brain, they are evident in neuronal avalanches—scale-invariant cascades of neuronal activity quantified by power laws. Avalanches manifest at the cellular level as cascades of neuronal groups that fire action potentials simultaneously. Such spatiotemporal synchronization is vital to theories on brain function yet avalanche synchronization is often underestimated when only a fraction of neurons is observed. Here, we investigate biases from fractional sampling within a balanced network of excitatory and inhibitory neurons with all-to-all connectivity and critical branching process dynamics. We focus on how mean avalanche size scales with avalanche duration. For parabolic avalanches, this scaling is quadratic, quantified by the scaling exponent, χ = 2, reflecting rapid spatial expansion of simultaneous neuronal firing over short durations. However, in networks sampled fractionally, χ is significantly lower. We demonstrate that applying temporal coarse-graining and increasing a minimum threshold for coincident firing restores χ = 2, even when as few as 0.1% of neurons are sampled. This correction crucially depends on the network being critical and fails for near sub- and supercritical branching dynamics. Using cellular 2-photon imaging, our approach robustly identifies χ = 2 over a wide parameter regime in ongoing neuronal activity from frontal cortex of awake mice. In contrast, the common ‘crackling noise’ approach fails to determine χ under similar sampling conditions at criticality. Our findings overcome scaling bias from fractional sampling and demonstrate rapid, spatiotemporal synchronization of neuronal assemblies consistent with scale-invariant, parabolic avalanches at criticality.

Abstract Silicon-based two-dimensional (2D) materials have garnered significant attention due to their unique properties and potential applications in electronics, optoelectronics, and other advanced technologies. Here, we present a comprehensive investigation of a novel silicon allotrope, Popsilicene (Pop-Si), derived from the structure of Popgraphene. Using density functional theory and ab initio molecular dynamics simulations, we explore the thermal stability, mechanical and electronic properties, and optical characteristics of Pop-Si. Our results demonstrate that Pop-Si exhibits good thermal stability at 1000 K. Electronic structure calculations reveal that Pop-Si is metallic, with a high density of states at the Fermi level. Furthermore, our analysis of the optical properties indicates that Pop-Si has pronounced UV–Vis optical activity, making it a promising candidate for optoelectronic devices. Mechanical property assessments show that Pop-Si has Young’s modulus ranging from 10 to 92 GPa and a Poisson’s ratio of 0.95. These results combined suggest its potential for practical applications.

Abstract Untargeted metabolomic analysis is a powerful tool used for the discovery of novel biomarkers. Chagas disease (CD), caused by Trypanosoma cruzi, is a neglected tropical disease that affects 6–7 million people with approximately 30% developing cardiac manifestations. The most significant clinical challenge lies in its long latency period after acute infection, and the lack of surrogate markers to predict disease progression or cure. In this cross-sectional study, we analyzed sera from 120 individuals divided into four groups: 31 indeterminate CD, 41 chronic chagasic cardiomyopathy (CCC), 18 Latin Americans with other cardiomyopathies and 30 healthy volunteers. Using a high-throughput panel of 986 metabolites, we identified three distinct profiles among individuals with cardiomyopathy, indeterminate CD and healthy volunteers. After a more stringent analysis, we identified some potential biomarkers. Among peptides, phenylacetylglutamine and fibrinopeptide B (1–13) exhibited an increasing trend from controls to ICD and CCC. Conversely, reduced levels of bilirubin and biliverdin alongside elevated urobilin correlated with disease progression. Finally, elevated levels of cystathionine, phenol glucuronide and vanillactate among amino acids distinguished CCC individuals from ICD and controls. Our novel exploratory study using metabolomics identified potential biomarker candidates, either alone or in combination that if confirmed, can be translated into clinical practice.

Abstract The safety of exercise practice by individuals with cardiovascular disease is due to a good clinical evaluation to identify patients with a higher risk of adverse events, thus the study of simple variables capable of predicting the rate of events during exercise is relevant and may provide better screening tools to be applied in the clinical practice. The study aimed to evaluate if clinical and physical parameters can predict the rate of adverse events during exercise-based cardiac rehabilitation. This was a cohort study that followed 73 patients during 24 sessions of exercise. Were registered 217 simple adverse events, at a rate of 2.5 events per hour of exercise. We found that higher adiposity, worse pulmonary function, lower functional capacity, and reduced heart rate variability were significant and acceptable predictors of adverse events during exercise, according to the ROC curve analysis. We state that these simple physical parameters may be useful for cardiac risk stratification in cardiac rehabilitation programs with low resources, contributing to its safety.

Abstract Flavonoids are compounds that result from the secondary metabolism of plants and play a crucial role in plant development and mitigating biotic and abiotic stresses. The highest levels of flavonoids are found in legumes such as soybean. Breeding programs aim to increase desirable traits, such as higher flavonoid contents and vigorous seeds. Soybeans are one of the richest sources of protein in the plant kingdom and the main source of flavonoid derivatives for human health. In view of this, the hypothesis of this study is based on the possibility that the concentration of isoflavones in soybean seeds contributes to the physiological quality of the seeds. The aim of this study was to analyze the content of flavonoids in soybean genotypes and their influence on the physiological quality of the seeds. Seeds from thirty-two soybean genotypes were obtained by carrying out a field experiment during the 2021/22 crop season. The experimental design was randomized blocks with four replications and thirty-two F3 soybean populations. The seeds obtained were subjected to germination, first germination counting, electrical conductivity and tetrazolium vigor and viability tests. After drying and milling the material from each genotype, liquid chromatography analysis was carried out to obtain flavonoids, performed at UPLC level. Data were submitted to analysis of variance and, when significant, the means were compared using the Scott-Knott test at 5% probability. The results found here show the occurrence of genotypes with higher amounts of flavonoids when compared to their peers. The flavonoid FLVD_G2 had the highest concentration and differed from the others. Thus, we can assume that the type and concentration of flavonoids does not influence the physiological quality of seeds from different soybean genotypes, but it does indirectly contribute to viability and vigor, since the genotypes with the highest FLVD_G2 levels had better FGC values. The findings indicate that there is a difference between the content of flavonoids in soybean genotypes, with a higher content of genistein. The content of flavonoids does not influence the physiological quality of seeds, but contributes to increasing viability and vigor.

Abstract Aedes aegypti and Aedes albopictus are the main vectors of arboviruses such as Dengue, Chikungunya and Zika, causing a major impact on global economic and public health. The main way to prevent these diseases is vector control, which is carried out through physical and biological methods, in addition to environmental management. Although chemical insecticides are the most effective strategy, they present some problems such as vector resistance and ecotoxicity. Recent research highlights the potential of the imidazolium salt "1-methyl-3-octadecylimidazolium chloride" (C18MImCl) as an innovative and environmentally friendly solution against Ae. aegypti. Despite its promising larvicidal activity, the mode of action of C18MImCl in mosquito cells and tissues remains unknown. This study aimed to investigate its impacts on Ae. aegypti larvae and three cell lines of Ae. aegypti and Ae. albopictus, comparing the cellular effects with those on human cells. Cell viability assays and histopathological analyses of treated larvae were conducted. Results revealed the imidazolium salt’s high selectivity (> 254) for mosquito cells over human cells. After salt ingestion, the mechanism of larval death involves toxic effects on midgut cells. This research marks the first description of an imidazolium salt's action on mosquito cells and midgut tissues, showcasing its potential for the development of a selective and sustainable strategy for vector control.

Abstract Imbalances in electrolyte concentrations can have severe consequences, but accurate and accessible measurements could improve patient outcomes. The current measurement method based on blood tests is accurate but invasive and time-consuming and is often unavailable for example in remote locations or an ambulance setting. In this paper, we explore the use of deep neural networks (DNNs) for regression tasks to accurately predict continuous electrolyte concentrations from electrocardiograms (ECGs), a quick and widely adopted tool. We analyze our DNN models on a novel dataset of over 290,000 ECGs across four major electrolytes and compare their performance with traditional machine learning models. For improved understanding, we also study the full spectrum from continuous predictions to a binary classification of extreme concentration levels. Finally, we investigate probabilistic regression approaches and explore uncertainty estimates for enhanced clinical usefulness. Our results show that DNNs outperform traditional models but model performance varies significantly across different electrolytes. While discretization leads to good classification performance, it does not address the original problem of continuous concentration level prediction. Probabilistic regression has practical potential, but our uncertainty estimates are not perfectly calibrated. Our study is therefore a first step towards developing an accurate and reliable ECG-based method for electrolyte concentration level prediction—a method with high potential impact within multiple clinical scenarios.