Explore the vast range of titles available.

Combining 3D printing with precursor-derived ceramic for fabricating electromagnetic (EM) wave-absorbing metamaterials has attracted great attention. This study presents a novel ultraviolet-curable polysiloxane precursor for digital light processing (DLP) 3D printing to fabricate ceramic parts with complex geometry, no cracks and linear shrinkage. Guiding with the principles of impedance matching, attenuation, and effective-medium theory, we design a cross-helix-array metamaterial model based on the complex permittivity constant of precursor-derived ceramics. The corresponding ceramic metamaterials can be successfully prepared by DLP printing and subsequent pyrolysis process, achieving a low reflection coefficient and a wide effective absorption bandwidth in the X-band even under high temperature. This is a general method that can be extended to other bands, which can be realized by merely adjusting the unit structure of metamaterials. This strategy provides a novel and effective avenue to achieve “target-design-fabricating” ceramic metamaterials, and it exposes the downstream applications of highly efficient and broad EM wave-absorbing materials and structures with great potential applications.

Backgrounds Osteosarcomas are one of the most common primary malignant tumors of bone. It primarily occurs in children and adolescents, with the second highest incidence among people over 50 years old. Although there were immense improvements in the survival of patients with osteosarcoma in the past 30 years, targetable mutations and agents of osteosarcomas still have been generally not satisfactory. Therefore, it is of great importance to further explore the highly specialized immune environment of bone, genes related to macrophage infiltration and potential therapeutic biomarkers and targets. Methods The 11 expression data sets of OS tissues and the 11 data sets of adjacent non-tumorous tissues available in the GEO database GSE126209 were used to conduct immune infiltration analysis. Then, through WGCNA analysis, we acquired the co-expression modules related to Mast cells activated and performed the GO and KEGG enrichment analysis. Next, we did the survival prognosis analysis and plotted a survival curve. Finally, we analyzed the COX multivariate regression of gene expression on clinical parameters and drew forest maps for visualization by the forest plot package. Results OS disease-related immune cell populations, mainly Mast cells activated, have higher cell content ( p  = 0.006) than the normal group. Then, we identified co-expression modules related to Mast cells activated. In sum, a total of 822 genes from the top three strongest positive correlation module MEbrown4, MEdarkslateblue and MEnavajowhite2 and the strongest negative correlation module MEdarkturquoise. From that, we identified nine genes with different levels in immune cell infiltration related to osteosarcoma, eight of which including SORBS2 , BAIAP2L2, ATAD2, CYGB, PAMR1, PSIP1, SNAPC3 and ZDHHC21 in their low abundance have higher disease-free survival probability than the group in their high abundances. Conclusion These results could assist clinicians to select targets for immunotherapies and individualize treatment strategies for patients with OS.

Spinal cord injury (SCI) is a serious traumatic disease of the central nervous system and leads to incomplete or complete loss of the body’s autonomous motor and sensory functions, seriously endangering human health. Recently, exosomes have been proposed as important substances in cell-to-cell interactions. Mesenchymal stem cell (MSC)-derived exosomes exert good therapeutic effects and play a crucial role in neurological damage repair. However, the detailed mechanisms underlying their effects remain unknown. Herein, we found that compared to SCI rats, those subjected to umbilical cord MSC (UC-MSC)-derived exosomes injection showed an improved motor ability. Nevertheless, the transcriptome of BV2 microglia in different treatment groups indicated that the action pathway of exosomes might be the NF-κB/MAPK pathway. Additionally, exosomes from UC-MSCs could inhibit P38, JNK, ERK, and P65 phosphorylation in BV2 microglia and SCI rat tissues. Moreover, exosomes could inhibit apoptosis and inflammatory reaction and reactive oxygen species (ROS) production of BV2 microglia in vitro and in vivo. In conclusion, UC-MSCs-derived exosomes might protect SCI in rats by inhibiting inflammatory response via the NF-κB/MAPK signaling pathway, representing novel treatment targets or approaches for SCI.

Background N6-methyl adenosine (m6A) is the most common reversible mRNA modification in eukaryotes implicated in key roles in various biological processes. The purpose of our analysis was to examine the association of ankylosing spondylitis (AS) with m6A methylation. Method We obtained 72 samples from the data set GSE73754, including 52 AS patients and 20 healthy people. We divided the samples into two groups: the experimental group and the control group, and then observed the differences of 26 m6A related genes in the two groups. We also analyzed the correlation between different m6A genes. We used a random forest tree model to screen seven m6A signature genes associated with AS to evaluate its prevalence. Next, the samples were classified according to the m6a content and differential genes. Immune analysis, gene ontology, and KEGG enrichment analyses were performed. Finally, we scored each sample with m6a and analyzed the relationship between different samples and inflammation-related factors. Results and conclusion In conclusion, we screened out AS-related genes and the nomogram showed that they were negatively correlated with the incidence of AS. And we found that AS may have some relationship with immunity. Our analysis results could provide further insights into the treatment of AS.

A computational spectrometer is a novel form of spectrometer powerful for portable in situ applications. In the encoding part of the computational spectrometer, filters with highly non-correlated properties are requisite for compressed sensing, which poses severe challenges for optical design and fabrication. In the reconstruction part of the computational spectrometer, conventional iterative reconstruction algorithms are featured with limited efficiency and accuracy, which hinders their application for real-time in situ measurements. This study proposes a neural network computational spectrometer trained by a small dataset with high-correlation optical filters. We aim to change the paradigm by which the accuracy of neural network computational spectrometers depends heavily on the amount of training data and the non-correlation property of optical filters. First, we propose a presumption about a distribution law for the common large training dataset, in which a unique widespread distribution law is shown when calculating the spectrum correlation. Based on that, we extract the original dataset according to the distribution probability and form a small training dataset. Then a fully connected neural network architecture is constructed to perform the reconstruction. After that, a group of thin film filters are introduced to work as the encoding layer. Then the neural network is trained by a small dataset under high-correlation filters and applied in simulation. Finally, the experiment is carried out and the result indicates that the neural network enabled by a small training dataset has performed very well with the thin film filters. This study may provide a reference for computational spectrometers based on high-correlation optical filters.

Spinal cord injury (SCI) is a devastating disorder featuring serious motor dysfunction and proprioceptive deficits due to central nervous system (CNS) damage. During its progression, macrophage-microglia (MM) cells are rapidly activated and play pivotal roles in the inflammatory response through various mechanisms. However, limited research has investigated the differential gene expression between the two groups before and after injury, and these important genes may therefore serve as potential biological diagnostic markers. Our study utilized the Gene Expression Omnibus (GEO) data for SCI-related MDEG identification. Key hub genes were screened using multiple machine learning(ML) algorithms. Their predictive potential was subsequently validated using independent datasets, and their association with immune cell infiltration was assessed. An model of SCI was established, and quantitative polymerase chain reaction (qPCR) experiments were conducted to verify our findings. Single-cell RNA sequencing identified 16 distinct cellular subpopulations, among which MM cells were split into four subsets according to functional characteristics. Enrichment analyses, including KEGG, GO, and GSEA, revealed that the MDEGs were closely linked to key biological processes in SCI. From 200 genes associated with critical WGCNA modules, three hub genes, including EMP3, GNGT2, and SGPL1, were identified through four ML algorithms as differentially expressed before and after injury. Predictive models based on these genes demonstrated strong performance in both internal training and external validation cohorts. Preliminary analysis of immune infiltration and gene-immune cell correlation suggests an association with M2 macrophages. Furthermore, modeling of post-injury inflammation confirmed the elevated expression of EMP3, GNGT2, and SGPL1. Our study identifies EMP3, GNGT2, and SGPL1 as potential diagnostic biomarkers associated with MM cells in SCI. The foregoing findings lay a theoretical basis for elucidating their biological roles and formulating future treatment strategies.

Background Large language models (LLMs) have shown promise in medical education and assessment, but their performance on highly specialized, context-specific exams remains under-explored. This study aimed to evaluate the accuracy of five advanced LLMs – Qwen3, DeepSeek-R1-0528, ChatGPT o3-pro, Gemini 2.5 pro, and Claude 4.0 – in answering oral and maxillofacial surgery multiple-choice questions, and compare their performance. Methods A total of 110 oral and maxillofacial surgery questions were collected from the 2013–2024 Chinese National Medical Licensing Examination (CNMLE), including 80 specialized knowledge questions and 30 case analysis questions. All questions were presented in Chinese to each LLM, and the correctness of each model’s first answer was recorded. We also attempted to verify the authenticity of any references provided in the models’ answers. Performance differences among models were analyzed with a significance level of P  < 0.05. Results All models achieved higher accuracy on knowledge-based questions than on case analysis questions. Qwen3 and DeepSeek showed no significant within-model accuracy difference between question types, whereas ChatGPT, Gemini, and Claude had significantly lower accuracy on case-based questions ( P  < 0.05). Across questions on tooth extraction, maxillofacial trauma/infection, and maxillofacial tumors, the models performed well (overall accuracy 87.3%–95.5%) with no significant inter-model differences. Qwen3 achieved the highest overall accuracy (95.5%), followed by DeepSeek (94.5%), ChatGPT (89.1%), and Gemini and Claude (87.3% each). However, 37.7%–43.9% of the references provided by the models were not verifiable, with no significant difference among the models. Conclusion LLMs demonstrated high accuracy on oral and maxillofacial surgery exam questions, indicating great potential as assistive tools in medical education and examinations. Nevertheless, their performance in complex case reasoning is notably lower and a substantial portion of their cited references are spurious. These findings highlight the need for further improvements in LLMs’ clinical reasoning capabilities and for reducing reference hallucinations before they can be reliably used in unsupervised clinical decision-making.

Enhancers are short non-coding DNA sequences outside of the target promoter regions that can be bound by specific proteins to increase a gene’s transcriptional activity, which has a crucial role in the spatiotemporal and quantitative regulation of gene expression. However, enhancers do not have a specific sequence motifs or structures, and their scattered distribution in the genome makes the identification of enhancers from human cell lines particularly challenging. Here we present a novel, stacked multivariate fusion framework called SMFM, which enables a comprehensive identification and analysis of enhancers from regulatory DNA sequences as well as their interpretation. Specifically, to characterize the hierarchical relationships of enhancer sequences, multi-source biological information and dynamic semantic information are fused to represent regulatory DNA enhancer sequences. Then, we implement a deep learning–based sequence network to learn the feature representation of the enhancer sequences comprehensively and to extract the implicit relationships in the dynamic semantic information. Ultimately, an ensemble machine learning classifier is trained based on the refined multi-source features and dynamic implicit relations obtained from the deep learning-based sequence network. Benchmarking experiments demonstrated that SMFM significantly outperforms other existing methods using several evaluation metrics. In addition, an independent test set was used to validate the generalization performance of SMFM by comparing it to other state-of-the-art enhancer identification methods. Moreover, we performed motif analysis based on the contribution scores of different bases of enhancer sequences to the final identification results. Besides, we conducted interpretability analysis of the identified enhancer sequences based on attention weights of EnhancerBERT, a fine-tuned BERT model that provides new insights into exploring the gene semantic information likely to underlie the discovered enhancers in an interpretable manner. Finally, in a human placenta study with 4,562 active distal gene regulatory enhancers, SMFM successfully exposed tissue-related placental development and the differential mechanism, demonstrating the generalizability and stability of our proposed framework.

Binary ethosome vesicles have been developed as flexible lipid vesicles for the enhanced physicochemical stability and skin delivery of drugs. This work aimed to prepare phloretin-loaded propylene glycol ethosomes (PHL-PGEs) to improve their stability, skin permeability and antioxidant activity. PHL-PGEs were prepared via the ethanol injection method and optimized using different weight ratios of ethanol to propylene glycol (PG). When the ethanol/PG mass ratio changed from 10:0 to 0:10, the encapsulation efficiency and stability of ethosomes increased. At a PHL concentration of 1mg/mL, the EE% was 89.42 ± 2.42 and the DL% was 4.21 ± 0.04, which exhibited their highest values. The encapsulation of the PHL in the PHL-PGEs was strengthened via XRD analysis and FTIR analysis. The results of the in vitro percutaneous permeability test demonstrated that the combined use of ethanol and PG exhibited a notable enhancement in skin permeability, and the skin retention of PHL-PGEs was 1.06 times that of PHL-ethosomes (PHL-Es) and 2.24 times that of the PHL solution. An in vitro antioxidant activity study indicated that solubility and antioxidant activity was potentiated via the nanoencapsulation of phloretin. Therefore, these results confirm the potential of this nanocarrier to enhance physicochemical stability, skin permeability and antioxidant activity.

In recent years, China's automobile industry has undergone rapid development, creating new opportunities for the auto loan industry. Currently, auto financing companies are actively seeking to expand their cooperation with banks. Therefore, improving the approval rate and scale of joint loan business is of significant practical importance. In this paper, we propose a Stacking‐based financial institution risk approval model and select the optimal stacking model by comparing its performance with other models. Additionally, we construct a bank approval model using deep learning techniques on a biased data set, with feature extraction performed using convolution neural networks (CNN) and feature‐based counterfactual augmentation used for balanced sampling. Finally, we optimize the model of the prediction of auto finance companies by selecting the optimal coefficients of loss function based on the features and results of the bank approval model. The proposed approach leads to an approximately 6% increase in the joint loan approval rate on the actual data set, as demonstrated by experimental results. In the context of joint loans, a bank approval model was built using a deep learning approach based on a dataset from an automotive finance company. Through methods such as label prediction, feature extraction, and loss function optimization, the performance of the stacked model for joint loans was enhanced.