Browse Articles

Study on physical clogging process and practical application of horizontal subsurface flow constructed wetland

A novel method for intelligent operation and maintenance of transformers using deep visual large model DETR + X and digital twin

Abstract To achieve real-time monitoring and intelligent maintenance of transformers, a framework based on deep vision and digital twin has been developed. An enhanced visual detection model, DETR + X, is proposed, implementing multidimensional sample data augmentation through Swin2SR and GAN networks. This model converts one-dimensional DGA data into three-dimensional feature images based on Gram angle fields, facilitating the transformation and fusion of heterogeneous modal information. The Pyramid Vision Transformer (PVT) is innovatively adopted as the backbone for image feature extraction, replacing the traditional ResNet structure. A Deformable Attention mechanism is employed to handle the complex spatial structure of multi-scale features. Testing results indicate that the improved DETR + X model performs well in transformer state recognition tasks, achieving a classification accuracy of 100% for DGA feature maps. In object detection tasks, it surpasses advanced models such as Faster R-CNN, RetinaNet, YOLOv8, and Deformable DETR in terms of overall mAP50 scores, particularly demonstrating significant enhancements in small object detection. Furthermore, the Llava-7b model, fine-tuned based on domain expertise, serves as an expert decision-making tool for transformer maintenance, providing accurate operational recommendations based on visual detection results. Finally, based on digital twin and inference models, a comprehensive platform has been developed to achieve real-time monitoring and intelligent maintenance of transformers.

Study on the simulation of bridge deformation in a mining subsidence area

The transcriptional repressor HEY2 regulates mitochondrial oxidative respiration to maintain cardiac homeostasis

Limitation of switching sensory information flow in flexible perceptual decision making

Exploring happiness factors with explainable ensemble learning in a global pandemic

Happiness is a state of contentment, joy, and fulfillment, arising from relationships, accomplishments, and inner peace, leading to well-being and positivity. The greatest happiness principle posits that morality is determined by pleasure, aiming for a society where individuals are content and free from suffering. While happiness factors vary, some are universally recognized. The World Happiness Report (WHR), published annually, includes data on ‘GDP per capita’, ‘social support’, ‘life expectancy’, ‘freedom to make life choices’, ‘generosity’, and ‘perceptions of corruption’. This paper predicts happiness scores using Machine Learning (ML), Deep Learning (DL), and ensemble ML and DL algorithms and examines the impact of individual variables on the happiness index. We also show the impact of COVID-19 pandemic on the happiness features. We design two ensemble ML and DL models using blending and stacking ensemble techniques, namely, Blending RGMLL, which combines Ridge Regression (RR), Gradient Boosting (GB), Multilayer Perceptron (MLP), Long Short-Term Memory (LSTM), and Linear Regression (LR), and Stacking LRGR, which combines LR, Random Forest (RF), GB, and RR. Among the trained models, Blending RGMLL demonstrates the highest predictive accuracy with R2 of 85%, MSE of 0.15, and RMSE of 0.38. We employ Explainable Artificial Intelligence (XAI) techniques to uncover changes in happiness indices, variable importance, and the impact of the COVID-19 pandemic on happiness. The study utilizes an open dataset from the WHR, covering 156 countries from 2018 to 2023. Our findings indicate that ‘GDP per capita’ is the most critical indicator of happiness score (HS), while ‘social support’ and ‘healthy life expectancy’ are also important features before and after the pandemic. However, during the pandemic, ‘social support’ emerged as the most important indicator, followed by ‘healthy life expectancy’ and ‘GDP per capita’, because social support is the prime necessity in the pandemic situation. The outcome of this research helps people understand the impact of these features on increasing the HS and provides guidelines on how happiness can be maintain during unwanted situations. Future research will explore advanced methods and include other related features with real-time monitoring for more comprehensive insights.

Effect and safety of sivelestat on acute severe pancreatitis with systemic inflammatory response syndrome: a retrospective study

Digital PCR detection of microsporidia in household pipe water of patients with microsporidial keratitis

Association between gut microbiota and short-chain fatty acids in children with obesity

The progression of basaltic–rhyolitic melt storage at Yellowstone Caldera

Engineering ultra-strong electron-phonon coupling and nonclassical electron transport in crystalline gold with nanoscale interfaces

AbstractElectrical resistivity in good metals, particularly noble metals such as gold (Au), silver (Ag), or copper, increases linearly with temperature (T) for T > ΘD, where ΘD is the Debye temperature. This is because the coupling (λ) between the electrons and the lattice vibrations, or phonons, in these metals is weak, with λ ~ 0.1−0.2. In this work, we outline a nanostructuring strategy of crystalline Au where this concept of metallic transport breaks down. We show that by embedding a distributed network of ultra-small Ag nanoparticles (AgNPs) of radius ~ 1–2 nm inside a crystalline Au shell, the electron-phonon interaction can be enhanced, with an effective λ as high as  ≈ 20. With increasing AgNP density, the electrical resistivity deviates from T-linearity and approaches a saturation to the Mott-Ioffe-Regel scale ρMIR ~ ha/e2 for both disorder (T → 0) and phonon (T ≫ ΘD)-dependent components of resistivity (here, a = 0.3 nm, is the lattice constant of Au).

Modulation of glymphatic system by visual circuit activation alleviates memory impairment and apathy in a mouse model of Alzheimer’s disease

Predicting learning achievement using ensemble learning with result explanation

Predicting learning achievement is a crucial strategy to address high dropout rates. However, existing prediction models often exhibit biases, limiting their accuracy. Moreover, the lack of interpretability in current machine learning methods restricts their practical application in education. To overcome these challenges, this research combines the strengths of various machine learning algorithms to design a robust model that performs well across multiple metrics, and uses interpretability analysis to elucidate the prediction results. This study introduces a predictive framework for learning achievement based on ensemble learning techniques. Specifically, six distinct machine learning models are utilized to establish a base learner, with logistic regression serving as the meta learner to construct an ensemble model for predicting learning achievement. The SHapley Additive exPlanation (SHAP) model is then employed to explain the prediction results. Through the experiments on XuetangX dataset, the effectiveness of the proposed model is verified. The proposed model outperforms traditional machine learning and deep learning model in terms of prediction accuracy. The results demonstrate that the ensemble learning-based predictive framework significantly outperforms traditional machine learning methods. Through feature importance analysis, the SHAP method enhances model interpretability and improves the reliability of the prediction results, enabling more personalized interventions to support students.

Complement classical and alternative pathway activation contributes to diabetic kidney disease progression: a glomerular proteomics on kidney biopsies

New formyl indole derivatives based on thiobarbituric acid and their nano-formulations; synthesis, characterization, parasitology and histopathology investigations

Assessment of changes in soil contact stress depending on tractor tire parameters

High-resolution genomic history of early medieval Europe

AbstractMany known and unknown historical events have remained below detection thresholds of genetic studies because subtle ancestry changes are challenging to reconstruct. Methods based on shared haplotypes1,2 and rare variants3,4 improve power but are not explicitly temporal and have not been possible to adopt in unbiased ancestry models. Here we develop Twigstats, an approach of time-stratified ancestry analysis that can improve statistical power by an order of magnitude by focusing on coalescences in recent times, while remaining unbiased by population-specific drift. We apply this framework to 1,556 available ancient whole genomes from Europe in the historical period. We are able to model individual-level ancestry using preceding genomes to provide high resolution. During the first half of the first millennium ce, we observe at least two different streams of Scandinavian-related ancestry expanding across western, central and eastern Europe. By contrast, during the second half of the first millennium ce, ancestry patterns suggest the regional disappearance or substantial admixture of these ancestries. In Scandinavia, we document a major ancestry influx by approximately 800 ce, when a large proportion of Viking Age individuals carried ancestry from groups related to central Europe not seen in individuals from the early Iron Age. Our findings suggest that time-stratified ancestry analysis can provide a higher-resolution lens for genetic history.

2.5-dimensional covalent organic frameworks

AbstractCovalently bonded crystalline substances with micropores have broad applications. Covalent organic frameworks (COFs) are representative of such substances. They have so far been classified into two-dimensional (2D) and three-dimensional (3D) COFs. 2D-COFs have planar shapes useful for broad purposes, but obtaining good crystals of 2D-COFs with sizes larger than 10 μm is significantly challenging, whereas yielding 3D-COFs with high crystallinity and larger sizes is easier. Here, we show COFs with 2.5-dimensional (2.5D) skeletons, which are microscopically constructed with 3D bonds but have macroscopically 2D planar shapes. The 2.5D-COFs shown herein achieve large single-crystal sizes above 0.1 mm and ultrahigh-density primary amines regularly allocated on and pointing perpendicular to the covalently-bonded network plane. Owing to the latter nature, the COFs are promising as CO2 adsorbents that can simultaneously achieve high CO2/N2 selectivity and low heat of adsorption, which are usually in a mutually exclusive relationship. 2.5D-COFs are expected to broaden the frontier and application of covalently bonded microporous crystalline systems.

Polyolefin reweaved ultra-micropore membrane for CO2 capture

Prognostic significance of total choline on in-vivo proton MR spectroscopy for prediction of late recurrence in patients with hormone receptor-positive, HER2-negative early breast cancer

Purpose In-vivo proton magnetic resonance spectroscopy (MRS) is a non-invasive method of analyzing choline metabolism that has been used to predict breast cancer prognosis. A strong choline peak may be a surrogate for aggressive tumor biology but its clinical relevance is unclear. The present study assessed whether total choline (tCho), as measured by proton MRS, can predict late recurrence in patients with hormone receptor (HR)-positive, HER2-negative early breast cancer. Methods The study cohort included 261 HR+/HER2- breast cancer patients who underwent diagnostic single-voxel proton MRS (3.0T scanner) prior to first-line surgery from March 2011 to July 2014. The relationships between tCho compound peak integral (tChoi) values and others prognostic factor were analyzed, as were the effects of tChoi on 10-year disease-free survival (DFS) and overall survival (OS). The clinical significance of tChoi was also analyzed using Harrell’s C-index. Results Mean tChoi in HR+/HER2- study group was 15.47 and we set the cut-off for tChoi at 15 for survival analysis. 10-year DFS differed significantly between tChoi <15 and ≥15 (p = 0.017), with differences differing significantly for late (5–10 years; p = 0.02) but not early (0–5 years; p = 0.323) recurrence. Cox regression analysis showed that tChoi was significantly predictive of 10-year DFS (p = 0.046, OR 2.69) and tended to be predictive of late recurrence (HR 4.36, p = 0.066). Harrell’s C-index showed that the Ki-67 index (AUC = 0.597) and lymphovascular invasion (AUC = 0.545) were also predictive of survival, with the addition of normalized tChoi improving the AUC to 0.622 (p = 0.014), indicating better predictive power. Conclusion tChoi determined by in vivo MRS was predictive of prognosis in patients with HR+/HER2- early breast cancer. This parameter may serve as a valuable, non-invasive tool to predict prognosis when combined with other known prognostic factors.