Chitosan(CS),a natural polymer derived from chitin found in the exoskeletons of crustaceans,has garnered significant interest in the pharmaceutical field due to its unique properties,including biocompatibility and bio...Chitosan(CS),a natural polymer derived from chitin found in the exoskeletons of crustaceans,has garnered significant interest in the pharmaceutical field due to its unique properties,including biocompatibility and biodegradability.In recent years,various studies have reported that CS can affect drug bioavailability,and interestingly,it works as an oral absorption enhancer and inhibitor.This review offers an in-depth analysis of the mechanisms underlying such a phenomenon and supports its application as a pharmaceutical excipient.CS enhances oral drug absorption through various mechanisms,such as interaction with the intestinal mucosa,tight junction modulation,inhibition of efflux transporters,enzyme inhibition,solubility and stability enhancement,and complexation.On the other side,CS exhibits the ability to inhibit the absorption of certain drugs by adsorbing to lipids and sterols,modulating bile acids and gut microbiota,altering drug-cell interaction at the polar interface,and mucus-mediated entrapment and interference.Future potential pharmaceutical research in this field includes elucidating the underneath absorption relevant mechanisms,rational use in formulations as excipient,exploring functional CS derivatives,and developing CS-based drug delivery systems.This comprehensive review highlights CS's versatile and significant role in enhancing and inhibiting oral drug absorption,providing insights into the complexities of drug delivery and the potential of CS to improve therapeutic outcomes.展开更多
Background:Long non-coding RNAs(lncRNAs)act as epigenetic regulators for tumor hallmarks.This investigation sought to probe the carcinogenic trait of PAN3-AS1 across pan-cancer comprehensively.Methods:We studied the d...Background:Long non-coding RNAs(lncRNAs)act as epigenetic regulators for tumor hallmarks.This investigation sought to probe the carcinogenic trait of PAN3-AS1 across pan-cancer comprehensively.Methods:We studied the diagnostic and prognostic features and the immune landscape of PAN3-AS1 across pan-cancer by bioinformatics approaches.The hierarchical regulatory networks governing PAN3-AS1 expression in colon cancer were explored via chromatin immunoprecipitation,luciferase activity assays,and RNA immunoprecipitation,etc.We screened drugs sensitive to WAP four-disulfide core domain 13(WFDC13)by virtual screening and molecular docking.Results:Single-cell transcriptomics demonstrated that a variety of immune populations abnormally expressed PAN3-AS1 beyond tumor cells.Integration of data from multiple databases revealed that PAN3-AS1 was highly expressed and associated with a bad prognosis in various malignancies.Notably,PAN3-AS1 expression was correlated with a suppressive immune microenvironment.Moreover,we observed poor immunotherapy efficacy when PAN3-AS1 was highly expressed in melanoma.In vitro assays and functional enrichment analysis revealed that PAN3-AS1 was associated with cell proliferation and the immune response in colon cancer.Our experiments confirmed that PAN3-AS1 facilitated WFDC13 expression through competitive binding to hsa-miR-423-5p in colon cancer.Moreover,the present paper illustrated that enhancer activity exerts an important modulatory ability for PAN3-AS1 expression.Conclusion:In short,PAN3-AS1 is a valuable biomarker for diagnosis and prognosis.PAN3-AS1 exhibits linkage to a cold tumor immune microenvironment(TME)and forecasts durable benefit from immunotherapy.Addressing the PAN3-AS1/miR-423-5p/WFDC13 axis might provide a novel option for improving immunotherapy efficacy in colon cancer.展开更多
Vertebrate axis patterning requires precise control of the differentiation of neuromesodermal progenitors(NMPs),which generate spinal cord(SC)and presomitic mesoderm(PSM).Previously,we identified a gastrula-premarked ...Vertebrate axis patterning requires precise control of the differentiation of neuromesodermal progenitors(NMPs),which generate spinal cord(SC)and presomitic mesoderm(PSM).Previously,we identified a gastrula-premarked posterior enhancer(p-Enh)that is essential for posterior tissue development by regulating somite and SC in organogenetic embryos,while its role in early NMPs cells remains elusive.Here,using a highly efficient in vitro differentiation system,we found that the genetic removal of p-Enh leads to the aberrantly up-regulated PSM-related genes during both PSM and SC differentiation.Time-resolved transcriptomic analysis and experimental characterization revealed the activated PSM transcriptomic signature arose from disorganized NMPs composition,with an over-representation of the T^(high)SOX2^(low)NMPs subtype.Besides,through a newly developed bioinformatic tool,ST-Pheno,which effectively bridges the in vitro samples to in vivo embryonic phenotypes within spatiotemporal context,we determined that the over-produced T^(high)SOX2^(low)NMPs subtype is predominantly enriched in the anterior primitive streak and adjacent mesoderm region at E7.5,which may disrupt the proper development of NMPs towards prospective PSM and SC,ultimately leading to the posterior development failure.In summary,this study demonstrates a critical role of p-Enh in regulating NMPs subtype composition,which will broaden the molecular understanding of mammalian embryogenesis.展开更多
Magnolol,a compound extracted from Magnolia officinalis,demonstrates potential efficacy in addressing metabolic dysfunction and cardiovascular diseases.Its biological activities encompass anti-inflammatory,antioxidant...Magnolol,a compound extracted from Magnolia officinalis,demonstrates potential efficacy in addressing metabolic dysfunction and cardiovascular diseases.Its biological activities encompass anti-inflammatory,antioxidant,anticoagulant,and anti-diabetic effects.Growth/differentiation factor-15(GDF-15),a member of the transforming growth factorβsuperfamily,is considered a potential therapeutic target for metabolic disorders.This study investigated the impact of magnolol on GDF-15 production and its underlying mechanism.The research examined the pharmacological effect of magnolol on GDF-15 expression in vitro and in vivo,and determined the involvement of endoplasmic reticulum(ER)stress signaling in this process.Luciferase reporter assays,chromatin immunoprecipitation,and in vitro DNA binding assays were employed to examine the regulation of GDF-15 by activating transcription factor 4(ATF4),CCAAT enhancer binding proteinγ(CEBPG),and CCCTC-binding factor(CTCF).The study also investigated the effect of magnolol and ATF4 on the activity of a putative enhancer located in the intron of the GDF-15 gene,as well as the influence of single nucleotide polymorphisms(SNPs)on magnolol and ATF4-induced transcription activity.Results demonstrated that magnolol triggers GDF-15 production in endothelial cells(ECs),hepatoma cell line G2(HepG2)and hepatoma cell line 3B(Hep3B)cell lines,and primary mouse hepatocytes.The cooperative binding of ATF4 and CEBPG upstream of the GDF-15 gene or the E1944285 enhancer located in the intron led to full-power transcription of the GDF-15 gene.SNP alleles were found to impact the magnolol and ATF4-induced transcription activity of GDF-15.In high-fat diet ApoE^(-/-)mice,administration of magnolol induced GDF-15 production and partially suppressed appetite through GDF-15.These findings suggest that magnolol regulates GDF-15 expression through priming of promoter and enhancer activity,indicating its potential as a drug for the treatment of metabolic disorders.展开更多
[Objectives]This study was conducted to investigate the effects of biochar-based fertilizer on high yield and quality of green pepper(Zanthoxylum armatum var.novemfolius)and soil nutrient status.[Methods]With green pe...[Objectives]This study was conducted to investigate the effects of biochar-based fertilizer on high yield and quality of green pepper(Zanthoxylum armatum var.novemfolius)and soil nutrient status.[Methods]With green pepper and pepper field soil as the research subjects,four treatments were set up,namely,conventional fertilization(T_(1)),conventional fertilization+biochar(T_(2)),biochar-based fertilizer 1(T_(3)),and biochar-based fertilizer 2(T_(4)).[Results]The application of biochar-based fertilizer increased green pepper yield by 9.37%-51.12%,with the order of increase being T_(4)>T_(3)>T_(2)>T_(1).In terms of soil nutrients,biochar-based fertilizer raised soil pH by 6.67%-53.33%,with the order of increase being T_(3)>T_(4)>T_(2)>T_(1).The initially strongly acidic soil gradually shifted to weakly acidic and approached neutral,indicating significantly improved soil acidity.The application of biochar-based fertilizer increased the contents of soil organic matter,available nitrogen,available phosphorus,available potassium,available copper,available zinc,available iron,and available manganese.It significantly enhanced green pepper yield,improved soil acidity,and elevated soil nutrient levels.Considering yield,nutrient uptake,and soil nutrient content,biochar-based fertilizer 2(T_(4))was identified as the optimal treatment.[Conclusions]This study provides a theoretical basis for improving green pepper yield and soil amendment.展开更多
The increasing challenges of environmental degradation,soil erosion,and climate change have driven interest in sustainable solutions like enhanced weathering(EW)and phytoremediation.Elephant Grass(Cenchrus purpureus),...The increasing challenges of environmental degradation,soil erosion,and climate change have driven interest in sustainable solutions like enhanced weathering(EW)and phytoremediation.Elephant Grass(Cenchrus purpureus),a fast-growing perennial species,shows promise as a bioaccumulator and agent for soil weathering.This study assessed the potential of C.purpureus to improve soil quality through heavy metal(HM)uptake and EW facilitation.A 60-day greenhouse pot experiment at Jackson State University evaluated plant performance in soils amended with metabasalt rock powder at 1:1 and 2:1 rock-to-soil ratios.Biomass,growth,and HM concentrations in roots and shoots were measured via ICP-MS after wet digestion.Soil pH and magnesium(Mg)release were also monitored to assess weathering and carbon drawdown.Results showed that C.purpureus accumulated more HMs in roots at higher amendment levels,while at lower levels,metals like As,Cd,and Cr were more translocated to shoots,enhancing phytoextraction potential.High treatment favored Fe and Al uptake,possibly reducing toxic metal accumulation in edible parts.Notably,C.purpureus contributed to the weathering of 38%of metabasalt rock,leading to a 42%increase in Mg release.With high biomass,HM tolerance,and weathering capacity,C.purpureus offers a sustainable strategy for soil remediation,improved soil health,and potential support for renewable energy systems.展开更多
Objective:In recent decades,studies have underscored nuclear proteins and signaling pathways in prostate cancer(PCa)development.Key biomarkers like Enhancer of zeste homolog 2(EZH2)and Forkhead box M1(FOXM1)are expres...Objective:In recent decades,studies have underscored nuclear proteins and signaling pathways in prostate cancer(PCa)development.Key biomarkers like Enhancer of zeste homolog 2(EZH2)and Forkhead box M1(FOXM1)are expressed in both healthy and malignant prostate cells.This study aimed to demonstrate the relationship between pathological characteristics,survival,recurrence,and tissue expression of EZH2 and FOXM1 in high-risk PCa patients.Methods:PCa tissues were used in a retrospective analysis that spanned from September 2009 to August 2019.Inclusion criteria comprised pathological tumor stage(pT)3 patients with positive surgical margins or tumor proximity to inked margins within 5 mm.After case selection,tissue slides were stained for EZH2 and FOXM1 antibodies,and Allred scores were calculated.Patients or relatives of deceased patients were contacted for signed agreements and disease follow-ups.Results:The pT3b,ductal carcinoma component,and moderate EZH2 expression were associated with relapse(odds ratio[OR]6.21,95%confidence interval[CI]1.41-27.27,p=0.016;OR 7.29,95%CI 1.03-51.43,p=0.046;OR 5.96,95%CI 1.09-32.48,p=0.039;respectively).The unilateral and bilateral seminal vesicle invasion increased the likelihood of recurrence by 9.98 times and 5.36 times,and the risk of death by around 9.78 times and 10.79 times,respectively.The pT3b was linked to higher death likelihood(OR 7.16,95%CI 1.38-37.23,p=0.019),while moderate EZH2 expression did not show statistical significance(OR 4.54,95%CI 0.87-23.60,p=0.072,marginally).Pathological regional lymph node stage(pN)1 had significantly higher probability of mortality than pN unknown(3.9%vs.27%,p<0.001).PCa in the neck and apex of the prostate gland increased death risk tenfold.Conclusion:Sufficient immunoexpression of EZH2,ductal carcinoma component,and neoplastic proliferation in the seminal vesicles,apex and neck of the prostate gland correlates with elevated risks of recurrence and mortality.Clinicians should use these criteria for appropriate patient referrals,and a multicenter trial could provide accurate classifications.展开更多
The existence of absorption and reflection of light underwater leads to problems such as color distortion and blue-green bias in underwater images.In this study,a depthwise separable convolution-based generative adver...The existence of absorption and reflection of light underwater leads to problems such as color distortion and blue-green bias in underwater images.In this study,a depthwise separable convolution-based generative adversarial network(GAN)algorithm was proposed.Taking GAN as the basic framework,it combined a depthwise separable convolution module,attention mechanism,and reconstructed convolution module to realize the enhancement of underwater degraded images.Multi-scale features were captured by the depthwise separable convolution module,and the attention mechanism was utilized to enhance attention to important features.The reconstructed convolution module further extracts and fuses local and global features.Experimental results showed that the algorithm performs well in improving the color bias and blurring of underwater images,with PSNR reaching 27.835,SSIM reaching 0.883,UIQM reaching 3.205,and UCIQE reaching 0.713.The enhanced image outperforms the comparison algorithm in both subjective and objective metrics.展开更多
Plp1-lineage Schwann cells(SCs)of peripheral nerve play a critical role in vascular remodeling and osteogenic differentiation during the early stage of bone healing,and the abnormal plasticity of SCs would jeopardize ...Plp1-lineage Schwann cells(SCs)of peripheral nerve play a critical role in vascular remodeling and osteogenic differentiation during the early stage of bone healing,and the abnormal plasticity of SCs would jeopardize the bone regeneration.However,how Plp1-lineage cells respond to injury and initiate the vascularized osteogenesis remains incompletely understood.Here,by employing single-cell transcriptional profiling combined with lineage-specific tracing models,we uncover that Plp1-lineage cells undergoing injury-induced glia-to-MSCs transition contributed to osteogenesis and revascularization in the initial stage of bone injury.Importantly,our data demonstrated that the Sonic hedgehog(Shh)signaling was responsible for the transition process initiation,which was strongly activated by c-Jun/SIRT6/BAF170 complex-driven Shh enhancers.Collectively,these findings depict an injuryspecific niche signal-mediated Plp1-lineage cells transition towards Gli1+MSCs and may be instructive for approaches to promote bone regeneration during aging or other bone diseases.展开更多
Background:Dysregulation of enhancer transcription occurs in multiple cancers.Enhancer RNAs(eRNAs)are transcribed products from enhancers that play critical roles in transcriptional control.Characterizing the genetic ...Background:Dysregulation of enhancer transcription occurs in multiple cancers.Enhancer RNAs(eRNAs)are transcribed products from enhancers that play critical roles in transcriptional control.Characterizing the genetic basis of eRNA expression may elucidate the molecular mechanisms underlying cancers.Methods:Initially,a comprehensive analysis of eRNA quantitative trait loci(eRNAQTLs)was performed in The Cancer Genome Atlas(TCGA),and functional features were characterized using multi-omics data.To establish the first eRNAQTL profiles for colorectal cancer(CRC)in China,epigenomic data were used to define active enhancers,which were subsequently integrated with transcription and genotyping data from 154 paired CRC samples.Finally,largescale case-control studies(34,585 cases and 69,544 controls)were conducted along with multipronged experiments to investigate the potential mechanisms by which candidate eRNAQTLs affect CRC risk.Results:A total of 300,112 eRNAQTLs were identified across 30 different cancer types,which exert their influence on eRNA transcription by modulating chromatin status,binding affinity to transcription factors and RNA-binding proteins.These eRNAQTLs were found to be significantly enriched in cancer risk loci,explaining a substantial proportion of cancer heritability.Additionally,tumor-specific eRNAQTLs exhibited high responsiveness to the development of cancer.Moreover,the target genes of these eRNAs were associated with dysregulated signaling pathways and immune cell infiltration in cancer,highlighting their potential as therapeutic targets.Furthermore,multiple ethnic population studies have confirmed that an eRNAQTL rs3094296-T variant decreases the risk of CRC in populations from China(OR=0.91,95%CI 0.88–0.95,P=2.92×10^(-7))and Europe(OR=0.92,95%CI 0.88–0.95,P=4.61×10^(-6)).Mechanistically,rs3094296 had an allele-specific effect on the transcription of the eRNA ENSR00000155786,which functioned as a transcriptional activator promoting the expression of its target gene SENP7.These two genes synergistically suppressed tumor cell proliferation.Our curated list of variants,genes,and drugs has been made available in CancereRNAQTL(http://canernaqtl.whu.edu.cn/#/)to serve as an informative resource for advancing this field.Conclusion:Our findings underscore the significance of eRNAQTLs in transcriptional regulation and disease heritability,pinpointing the potential of eRNA-based therapeutic strategies in cancers.展开更多
Medical imaging is essential in modern health care,allowing accurate diagnosis and effective treatment planning.These images,however,often demonstrate low contrast,noise,and brightness distortion that reduce their dia...Medical imaging is essential in modern health care,allowing accurate diagnosis and effective treatment planning.These images,however,often demonstrate low contrast,noise,and brightness distortion that reduce their diagnostic reliability.This review presents a structured and comprehensive analysis of advanced histogram equalization(HE)-based techniques for medical image enhancement.Our review methodology encompasses:(1)classical HE approaches and related limitations in medical domains;(2)adaptive schemes like Adaptive Histogram Equalization(AHE)and Contrast Limited Adaptive Histogrma Equalization(CLAHE)and their advance variants;(3)brightnesspreserving schemes like BBHE and MMBEBHE and related algorithms;(4)dynamic and recursive histogram equalization methods incorporating DHE and RMSHE;(5)fuzzy logic-based enhancement methodologies addressing uncertainty and noise in medical images;and(6)hybrid optimization methodologies through the application of metaheuristic algorithms(World Cup Optimization,Particle Swarm Optimization,Genetic Algorithms,along with histogram-based methodologies.)There is also a comparative discussion given based on contrast improvement,image brightness preservation,noise management,and computational efficiency.Such advancements have better capabilities of improving image quality,which is more important for improved diagnosis and image analysis.展开更多
Let G=(V,E)be a connected graph.For an integer h≥0,a subset F■V(G)(resp.F■E(G))of G,if any,is called an h-restricted vertex cut(resp.h-restricted edge cut)of G,if G-F is disconnected and every vertex in G-F has at ...Let G=(V,E)be a connected graph.For an integer h≥0,a subset F■V(G)(resp.F■E(G))of G,if any,is called an h-restricted vertex cut(resp.h-restricted edge cut)of G,if G-F is disconnected and every vertex in G-F has at least h neighbors.The cardinality of a minimum h-restricted vertex-cut(resp.h-restricted edge cut)of G is the h-restricted connectivity(resp.h-restricted edge connectivity)of G,and denoted by κ^(h)(G)(resp.λ^(h)(G)).The enhanced hypercube Q_(n,κ)(1≤k≤n)is a variant of the hypercube Q_(n).In this paper,we consider the h-restricted connectivity of Q_(n,κ) for 2≤k≤n-1.Our main results are as follows:(1)κ^(h)(Q_(n,κ))=2^(h)(n-h+1)for 4≤k≤n-1 and 0≤h≤n-3,λ^(h)(Q_(n,κ))=2^(h)(n-h+1)for 2≤k≤n-1 and 0≤h≤n-2.(2)κ^(h)(Q_(n,3))=2^(h-1)(n-h+1)for n≥5 and 4≤h≤n-1,κ^(h)(Q_(n,2))=2^(h-1)(n-h+1)for n≥4 and 3≤h≤n-1.(3)κ^(3)(Q_(n,3))=6n-16 for n≥5,κ^(2)(Q_(n,3))=4n-8 for n≥4 and κ^(2)(Q_(n,2))=3n-5 for n≥3,κ^(1)(Q_(n,3))=2n and κ^(3)(Q_(n,2))=2n-2 for n≥3.展开更多
Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approach...Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approaches,while effective in global illumination modeling,often struggle to simultaneously suppress noise and preserve structural details,especially under heterogeneous lighting.Furthermore,misalignment between luminance and color channels introduces additional challenges to accurate enhancement.In response to the aforementioned difficulties,we introduce a single-stage framework,M2ATNet,using the multi-scale multi-attention and Transformer architecture.First,to address the problems of texture blurring and residual noise,we design a multi-scale multi-attention denoising module(MMAD),which is applied separately to the luminance and color channels to enhance the structural and texture modeling capabilities.Secondly,to solve the non-alignment problem of the luminance and color channels,we introduce the multi-channel feature fusion Transformer(CFFT)module,which effectively recovers the dark details and corrects the color shifts through cross-channel alignment and deep feature interaction.To guide the model to learn more stably and efficiently,we also fuse multiple types of loss functions to form a hybrid loss term.We extensively evaluate the proposed method on various standard datasets,including LOL-v1,LOL-v2,DICM,LIME,and NPE.Evaluation in terms of numerical metrics and visual quality demonstrate that M2ATNet consistently outperforms existing advanced approaches.Ablation studies further confirm the critical roles played by the MMAD and CFFT modules to detail preservation and visual fidelity under challenging illumination-deficient environments.展开更多
Weakly Supervised Semantic Segmentation(WSSS),which relies only on image-level labels,has attracted significant attention for its cost-effectiveness and scalability.Existing methods mainly enhance inter-class distinct...Weakly Supervised Semantic Segmentation(WSSS),which relies only on image-level labels,has attracted significant attention for its cost-effectiveness and scalability.Existing methods mainly enhance inter-class distinctions and employ data augmentation to mitigate semantic ambiguity and reduce spurious activations.However,they often neglect the complex contextual dependencies among image patches,resulting in incomplete local representations and limited segmentation accuracy.To address these issues,we propose the Context Patch Fusion with Class Token Enhancement(CPF-CTE)framework,which exploits contextual relations among patches to enrich feature repre-sentations and improve segmentation.At its core,the Contextual-Fusion Bidirectional Long Short-Term Memory(CF-BiLSTM)module captures spatial dependencies between patches and enables bidirectional information flow,yield-ing a more comprehensive understanding of spatial correlations.This strengthens feature learning and segmentation robustness.Moreover,we introduce learnable class tokens that dynamically encode and refine class-specific semantics,enhancing discriminative capability.By effectively integrating spatial and semantic cues,CPF-CTE produces richer and more accurate representations of image content.Extensive experiments on PASCAL VOC 2012 and MS COCO 2014 validate that CPF-CTE consistently surpasses prior WSSS methods.展开更多
The development of synthetic hybrid biological systems integrating photosynthetic organisms with organic-abiotic functional materials holds significant promise for enhancing photosynthetic processes.The artificial reg...The development of synthetic hybrid biological systems integrating photosynthetic organisms with organic-abiotic functional materials holds significant promise for enhancing photosynthetic processes.The artificial regulation of the state transition between photosystem I(PSI)and photosystem II(PSII)represents a strategic and promising approach for improving the efficiency of natural photosynthesis.In this study,we demonstrate that poly(benzimidazolium-phenylthiophene)(CP4)featuring a flexible cationic backbone exhibits superior ultraviolet light-harvesting capability.The polymer CP4 enhanced PSI activity in Chlorella pyrenoidosa(C.pyrenoidosa),subsequently promoting PSII activity and augmenting overall photosynthetic performance.During light-dependent reactions,CP4 significantly accelerated photosynthetic electron transfer,resulting in a 330%increase in the production of oxygen and 93%and 96%increases in the ATP and NADPH contents,respectively.In the context of dark reactions,CP4 facilitated the conversion and utilization of light energy,leading to a 6%increase in both carbohydrate and protein contents.These findings indicate that synthetic light-harvesting polymer materials exhibit considerable application potential in the field of biomass production through enhancement of natural photosynthetic efficiency.展开更多
Aiming at the problem of insufficient recognition of implicit variants by existing Chinese sensitive text detection methods,this paper proposes the IPKE-MoE framework,which consists of three parts,namely,a sensitive w...Aiming at the problem of insufficient recognition of implicit variants by existing Chinese sensitive text detection methods,this paper proposes the IPKE-MoE framework,which consists of three parts,namely,a sensitive word variant extraction framework,a sensitive word variant knowledge enhancement layer and a mixture-of-experts(MoE)classification layer.First,sensitive word variants are precisely extracted through dynamic iterative prompt templates and the context-aware capabilities of Large Language Models(LLMs).Next,the extracted variants are used to construct a knowledge enhancement layer for sensitive word variants based on RoCBert models.Specifically,after locating variants via n-gram algorithms,variant types are mapped to embedding vectors and fused with original word vectors.Finally,a mixture-of-experts(MoE)classification layer is designed(sensitive word,sentiment,and semantic experts),which decouples the relationship between sensitiveword existence and text toxicity throughmultiple experts.This framework effectively combines the comprehension ability of Large Language Models(LLMs)with the discriminative ability of smaller models.Our two experiments demonstrate that the sensitive word variant extraction framework based on dynamically iterated prompt templates outperforms other baseline prompt templates.TheRoCBert models incorporating the sensitive word variant knowledge enhancement layer and a mixture-of-experts(MoE)classification layer achieve superior classification performance compared to other baselines.展开更多
Low-light image enhancement aims to improve the visibility of severely degraded images captured under insufficient illumination,alleviating the adverse effects of illumination degradation on image quality.Traditional ...Low-light image enhancement aims to improve the visibility of severely degraded images captured under insufficient illumination,alleviating the adverse effects of illumination degradation on image quality.Traditional Retinex-based approaches,inspired by human visual perception of brightness and color,decompose an image into illumination and reflectance components to restore fine details.However,their limited capacity for handling noise and complex lighting conditions often leads to distortions and artifacts in the enhanced results,particularly under extreme low-light scenarios.Although deep learning methods built upon Retinex theory have recently advanced the field,most still suffer frominsufficient interpretability and sub-optimal enhancement performance.This paper presents RetinexWT,a novel framework that tightly integrates classical Retinex theory with modern deep learning.Following Retinex principles,RetinexWT employs wavelet transforms to estimate illumination maps for brightness adjustment.A detail-recovery module that synergistically combines Vision Transformer(ViT)and wavelet transforms is then introduced to guide the restoration of lost details,thereby improving overall image quality.Within the framework,wavelet decomposition splits input features into high-frequency and low-frequency components,enabling scale-specific processing of global illumination/color cues and fine textures.Furthermore,a gating mechanism selectively fuses down-sampled and up-sampled features,while an attention-based fusion strategy enhances model interpretability.Extensive experiments on the LOL dataset demonstrate that RetinexWT surpasses existing Retinex-oriented deeplearning methods,achieving an average Peak Signal-to-Noise Ratio(PSNR)improvement of 0.22 dB over the current StateOfTheArt(SOTA),thereby confirming its superiority in low-light image enhancement.Code is available at https://github.com/CHEN-hJ516/RetinexWT(accessed on 14 October 2025).展开更多
Underwater images often affect the effectiveness of underwater visual tasks due to problems such as light scattering,color distortion,and detail blurring,limiting their application performance.Existing underwater imag...Underwater images often affect the effectiveness of underwater visual tasks due to problems such as light scattering,color distortion,and detail blurring,limiting their application performance.Existing underwater image enhancement methods,although they can improve the image quality to some extent,often lead to problems such as detail loss and edge blurring.To address these problems,we propose FENet,an efficient underwater image enhancement method.FENet first obtains three different scales of images by image downsampling and then transforms them into the frequency domain to extract the low-frequency and high-frequency spectra,respectively.Then,a distance mask and a mean mask are constructed based on the distance and magnitude mean for enhancing the high-frequency part,thus improving the image details and enhancing the effect by suppressing the noise in the low-frequency part.Affected by the light scattering of underwater images and the fact that some details are lost if they are directly reduced to the spatial domain after the frequency domain operation.For this reason,we propose a multi-stage residual feature aggregation module,which focuses on detail extraction and effectively avoids information loss caused by global enhancement.Finally,we combine the edge guidance strategy to further enhance the edge details of the image.Experimental results indicate that FENet outperforms current state-of-the-art underwater image enhancement methods in quantitative and qualitative evaluations on multiple publicly available datasets.展开更多
基金financially supported by National Key Research and Development Program of China (No.2021YFD1800900)National Natural Science Foundation of China (No.82073790)+2 种基金Special Fund for Youth Team of Southwest University (No.SWUXJLJ202306)Chongqing Science and Technology Commission (Nos.CSTB2022TIAD-LUX0001,CSTB2023NSCQ-JQX0002)Innovation Research 2035 Pilot Plan of Southwest University (No.SWUXDPY22007)。
文摘Chitosan(CS),a natural polymer derived from chitin found in the exoskeletons of crustaceans,has garnered significant interest in the pharmaceutical field due to its unique properties,including biocompatibility and biodegradability.In recent years,various studies have reported that CS can affect drug bioavailability,and interestingly,it works as an oral absorption enhancer and inhibitor.This review offers an in-depth analysis of the mechanisms underlying such a phenomenon and supports its application as a pharmaceutical excipient.CS enhances oral drug absorption through various mechanisms,such as interaction with the intestinal mucosa,tight junction modulation,inhibition of efflux transporters,enzyme inhibition,solubility and stability enhancement,and complexation.On the other side,CS exhibits the ability to inhibit the absorption of certain drugs by adsorbing to lipids and sterols,modulating bile acids and gut microbiota,altering drug-cell interaction at the polar interface,and mucus-mediated entrapment and interference.Future potential pharmaceutical research in this field includes elucidating the underneath absorption relevant mechanisms,rational use in formulations as excipient,exploring functional CS derivatives,and developing CS-based drug delivery systems.This comprehensive review highlights CS's versatile and significant role in enhancing and inhibiting oral drug absorption,providing insights into the complexities of drug delivery and the potential of CS to improve therapeutic outcomes.
基金supported by the Shandong Province Medical and Health Technology Project(202303111442).
文摘Background:Long non-coding RNAs(lncRNAs)act as epigenetic regulators for tumor hallmarks.This investigation sought to probe the carcinogenic trait of PAN3-AS1 across pan-cancer comprehensively.Methods:We studied the diagnostic and prognostic features and the immune landscape of PAN3-AS1 across pan-cancer by bioinformatics approaches.The hierarchical regulatory networks governing PAN3-AS1 expression in colon cancer were explored via chromatin immunoprecipitation,luciferase activity assays,and RNA immunoprecipitation,etc.We screened drugs sensitive to WAP four-disulfide core domain 13(WFDC13)by virtual screening and molecular docking.Results:Single-cell transcriptomics demonstrated that a variety of immune populations abnormally expressed PAN3-AS1 beyond tumor cells.Integration of data from multiple databases revealed that PAN3-AS1 was highly expressed and associated with a bad prognosis in various malignancies.Notably,PAN3-AS1 expression was correlated with a suppressive immune microenvironment.Moreover,we observed poor immunotherapy efficacy when PAN3-AS1 was highly expressed in melanoma.In vitro assays and functional enrichment analysis revealed that PAN3-AS1 was associated with cell proliferation and the immune response in colon cancer.Our experiments confirmed that PAN3-AS1 facilitated WFDC13 expression through competitive binding to hsa-miR-423-5p in colon cancer.Moreover,the present paper illustrated that enhancer activity exerts an important modulatory ability for PAN3-AS1 expression.Conclusion:In short,PAN3-AS1 is a valuable biomarker for diagnosis and prognosis.PAN3-AS1 exhibits linkage to a cold tumor immune microenvironment(TME)and forecasts durable benefit from immunotherapy.Addressing the PAN3-AS1/miR-423-5p/WFDC13 axis might provide a novel option for improving immunotherapy efficacy in colon cancer.
基金supported in part by the National Key Basic Research and Development Program of China(2025YFE0200600,2018YFA0800100,2019YFA0801402)the Major Project of Guangzhou National Laboratory(GZNL2025C02014,GZNL2023A02005)+1 种基金the National Natural Science Foundation of China(32130030,32470866,31900454)the Union Project by Guangzhou National Laboratory and State Key Laboratory of Respiratory Disease,Guangzhou Medical University(GZNL2024B01007).
文摘Vertebrate axis patterning requires precise control of the differentiation of neuromesodermal progenitors(NMPs),which generate spinal cord(SC)and presomitic mesoderm(PSM).Previously,we identified a gastrula-premarked posterior enhancer(p-Enh)that is essential for posterior tissue development by regulating somite and SC in organogenetic embryos,while its role in early NMPs cells remains elusive.Here,using a highly efficient in vitro differentiation system,we found that the genetic removal of p-Enh leads to the aberrantly up-regulated PSM-related genes during both PSM and SC differentiation.Time-resolved transcriptomic analysis and experimental characterization revealed the activated PSM transcriptomic signature arose from disorganized NMPs composition,with an over-representation of the T^(high)SOX2^(low)NMPs subtype.Besides,through a newly developed bioinformatic tool,ST-Pheno,which effectively bridges the in vitro samples to in vivo embryonic phenotypes within spatiotemporal context,we determined that the over-produced T^(high)SOX2^(low)NMPs subtype is predominantly enriched in the anterior primitive streak and adjacent mesoderm region at E7.5,which may disrupt the proper development of NMPs towards prospective PSM and SC,ultimately leading to the posterior development failure.In summary,this study demonstrates a critical role of p-Enh in regulating NMPs subtype composition,which will broaden the molecular understanding of mammalian embryogenesis.
基金supported by the National Natural Science Foundation of China(Nos.82171552 and 82170479)the Natural Science Foundation of Shanghai Ctiy(No.21ZR1457500)the Science and Technology Bureau of Shanghai Putuo District(No.ptkwws202102).
文摘Magnolol,a compound extracted from Magnolia officinalis,demonstrates potential efficacy in addressing metabolic dysfunction and cardiovascular diseases.Its biological activities encompass anti-inflammatory,antioxidant,anticoagulant,and anti-diabetic effects.Growth/differentiation factor-15(GDF-15),a member of the transforming growth factorβsuperfamily,is considered a potential therapeutic target for metabolic disorders.This study investigated the impact of magnolol on GDF-15 production and its underlying mechanism.The research examined the pharmacological effect of magnolol on GDF-15 expression in vitro and in vivo,and determined the involvement of endoplasmic reticulum(ER)stress signaling in this process.Luciferase reporter assays,chromatin immunoprecipitation,and in vitro DNA binding assays were employed to examine the regulation of GDF-15 by activating transcription factor 4(ATF4),CCAAT enhancer binding proteinγ(CEBPG),and CCCTC-binding factor(CTCF).The study also investigated the effect of magnolol and ATF4 on the activity of a putative enhancer located in the intron of the GDF-15 gene,as well as the influence of single nucleotide polymorphisms(SNPs)on magnolol and ATF4-induced transcription activity.Results demonstrated that magnolol triggers GDF-15 production in endothelial cells(ECs),hepatoma cell line G2(HepG2)and hepatoma cell line 3B(Hep3B)cell lines,and primary mouse hepatocytes.The cooperative binding of ATF4 and CEBPG upstream of the GDF-15 gene or the E1944285 enhancer located in the intron led to full-power transcription of the GDF-15 gene.SNP alleles were found to impact the magnolol and ATF4-induced transcription activity of GDF-15.In high-fat diet ApoE^(-/-)mice,administration of magnolol induced GDF-15 production and partially suppressed appetite through GDF-15.These findings suggest that magnolol regulates GDF-15 expression through priming of promoter and enhancer activity,indicating its potential as a drug for the treatment of metabolic disorders.
基金Supported by General Project of Chongqing Technology Innovation and Application Demonstration(Social People s Livelihood Category)(cstc2018jscx-msybX0215)Horizontal Project of Chongqing Agricultural Technology Extension Station(WLHX-2021-0113).
文摘[Objectives]This study was conducted to investigate the effects of biochar-based fertilizer on high yield and quality of green pepper(Zanthoxylum armatum var.novemfolius)and soil nutrient status.[Methods]With green pepper and pepper field soil as the research subjects,four treatments were set up,namely,conventional fertilization(T_(1)),conventional fertilization+biochar(T_(2)),biochar-based fertilizer 1(T_(3)),and biochar-based fertilizer 2(T_(4)).[Results]The application of biochar-based fertilizer increased green pepper yield by 9.37%-51.12%,with the order of increase being T_(4)>T_(3)>T_(2)>T_(1).In terms of soil nutrients,biochar-based fertilizer raised soil pH by 6.67%-53.33%,with the order of increase being T_(3)>T_(4)>T_(2)>T_(1).The initially strongly acidic soil gradually shifted to weakly acidic and approached neutral,indicating significantly improved soil acidity.The application of biochar-based fertilizer increased the contents of soil organic matter,available nitrogen,available phosphorus,available potassium,available copper,available zinc,available iron,and available manganese.It significantly enhanced green pepper yield,improved soil acidity,and elevated soil nutrient levels.Considering yield,nutrient uptake,and soil nutrient content,biochar-based fertilizer 2(T_(4))was identified as the optimal treatment.[Conclusions]This study provides a theoretical basis for improving green pepper yield and soil amendment.
文摘The increasing challenges of environmental degradation,soil erosion,and climate change have driven interest in sustainable solutions like enhanced weathering(EW)and phytoremediation.Elephant Grass(Cenchrus purpureus),a fast-growing perennial species,shows promise as a bioaccumulator and agent for soil weathering.This study assessed the potential of C.purpureus to improve soil quality through heavy metal(HM)uptake and EW facilitation.A 60-day greenhouse pot experiment at Jackson State University evaluated plant performance in soils amended with metabasalt rock powder at 1:1 and 2:1 rock-to-soil ratios.Biomass,growth,and HM concentrations in roots and shoots were measured via ICP-MS after wet digestion.Soil pH and magnesium(Mg)release were also monitored to assess weathering and carbon drawdown.Results showed that C.purpureus accumulated more HMs in roots at higher amendment levels,while at lower levels,metals like As,Cd,and Cr were more translocated to shoots,enhancing phytoextraction potential.High treatment favored Fe and Al uptake,possibly reducing toxic metal accumulation in edible parts.Notably,C.purpureus contributed to the weathering of 38%of metabasalt rock,leading to a 42%increase in Mg release.With high biomass,HM tolerance,and weathering capacity,C.purpureus offers a sustainable strategy for soil remediation,improved soil health,and potential support for renewable energy systems.
文摘Objective:In recent decades,studies have underscored nuclear proteins and signaling pathways in prostate cancer(PCa)development.Key biomarkers like Enhancer of zeste homolog 2(EZH2)and Forkhead box M1(FOXM1)are expressed in both healthy and malignant prostate cells.This study aimed to demonstrate the relationship between pathological characteristics,survival,recurrence,and tissue expression of EZH2 and FOXM1 in high-risk PCa patients.Methods:PCa tissues were used in a retrospective analysis that spanned from September 2009 to August 2019.Inclusion criteria comprised pathological tumor stage(pT)3 patients with positive surgical margins or tumor proximity to inked margins within 5 mm.After case selection,tissue slides were stained for EZH2 and FOXM1 antibodies,and Allred scores were calculated.Patients or relatives of deceased patients were contacted for signed agreements and disease follow-ups.Results:The pT3b,ductal carcinoma component,and moderate EZH2 expression were associated with relapse(odds ratio[OR]6.21,95%confidence interval[CI]1.41-27.27,p=0.016;OR 7.29,95%CI 1.03-51.43,p=0.046;OR 5.96,95%CI 1.09-32.48,p=0.039;respectively).The unilateral and bilateral seminal vesicle invasion increased the likelihood of recurrence by 9.98 times and 5.36 times,and the risk of death by around 9.78 times and 10.79 times,respectively.The pT3b was linked to higher death likelihood(OR 7.16,95%CI 1.38-37.23,p=0.019),while moderate EZH2 expression did not show statistical significance(OR 4.54,95%CI 0.87-23.60,p=0.072,marginally).Pathological regional lymph node stage(pN)1 had significantly higher probability of mortality than pN unknown(3.9%vs.27%,p<0.001).PCa in the neck and apex of the prostate gland increased death risk tenfold.Conclusion:Sufficient immunoexpression of EZH2,ductal carcinoma component,and neoplastic proliferation in the seminal vesicles,apex and neck of the prostate gland correlates with elevated risks of recurrence and mortality.Clinicians should use these criteria for appropriate patient referrals,and a multicenter trial could provide accurate classifications.
文摘The existence of absorption and reflection of light underwater leads to problems such as color distortion and blue-green bias in underwater images.In this study,a depthwise separable convolution-based generative adversarial network(GAN)algorithm was proposed.Taking GAN as the basic framework,it combined a depthwise separable convolution module,attention mechanism,and reconstructed convolution module to realize the enhancement of underwater degraded images.Multi-scale features were captured by the depthwise separable convolution module,and the attention mechanism was utilized to enhance attention to important features.The reconstructed convolution module further extracts and fuses local and global features.Experimental results showed that the algorithm performs well in improving the color bias and blurring of underwater images,with PSNR reaching 27.835,SSIM reaching 0.883,UIQM reaching 3.205,and UCIQE reaching 0.713.The enhanced image outperforms the comparison algorithm in both subjective and objective metrics.
基金supported by the National Natural Science Foundation of China(grants 81970910 and 82370931)Jiangsu Province Capability Improvement Project through Science,Technology and Education-Jiangsu Provincial Research Hospital Cultivation Unit(YJXYYJSDW4)Jiangsu Provincial Medical Innovation Center(CXZX202227).
文摘Plp1-lineage Schwann cells(SCs)of peripheral nerve play a critical role in vascular remodeling and osteogenic differentiation during the early stage of bone healing,and the abnormal plasticity of SCs would jeopardize the bone regeneration.However,how Plp1-lineage cells respond to injury and initiate the vascularized osteogenesis remains incompletely understood.Here,by employing single-cell transcriptional profiling combined with lineage-specific tracing models,we uncover that Plp1-lineage cells undergoing injury-induced glia-to-MSCs transition contributed to osteogenesis and revascularization in the initial stage of bone injury.Importantly,our data demonstrated that the Sonic hedgehog(Shh)signaling was responsible for the transition process initiation,which was strongly activated by c-Jun/SIRT6/BAF170 complex-driven Shh enhancers.Collectively,these findings depict an injuryspecific niche signal-mediated Plp1-lineage cells transition towards Gli1+MSCs and may be instructive for approaches to promote bone regeneration during aging or other bone diseases.
基金supported by the National Science Fund for Excellent Young Scholars(NSFC-82322058)the Program of National Natural Science Foundation of China(NSFC-82103929,NSFC-82273713)+10 种基金the Young Elite Scientists Sponsorship Program by CAST(2022QNRC001)the National Science Fund for Distinguished Young Scholars of Hubei Province of China(2023AFA046)the Fundamental Research Funds for the Central Universities(WHU:2042022kf1205)Fundamental Research Funds for the Central Universities(WHU:2042022kf1031)for Ying Zhuthe Fundamental Research Funds for the Central Universities(2042022rc0026,2042023kf1005)for Xiao-Ping Miaothe Knowledge Innovation Program of Wuhan(whkxjsj011,2023020201010073)for Jian-Bo Tianthe Science and Technology Innovation Seed Fund of Zhongnan Hospital of Wuhan University(znpy2019060)for Yong-Chang Weithe Distinguished Young Scholars of China(NSFC-81925032)the Key Program of National Natural Science Foundation of China(NSFC-82130098)the Youth Program of National Natural Science Foundation of China(NSFC-82003547)the Program of Health Commission of Hubei Province(WJ2023M045)。
文摘Background:Dysregulation of enhancer transcription occurs in multiple cancers.Enhancer RNAs(eRNAs)are transcribed products from enhancers that play critical roles in transcriptional control.Characterizing the genetic basis of eRNA expression may elucidate the molecular mechanisms underlying cancers.Methods:Initially,a comprehensive analysis of eRNA quantitative trait loci(eRNAQTLs)was performed in The Cancer Genome Atlas(TCGA),and functional features were characterized using multi-omics data.To establish the first eRNAQTL profiles for colorectal cancer(CRC)in China,epigenomic data were used to define active enhancers,which were subsequently integrated with transcription and genotyping data from 154 paired CRC samples.Finally,largescale case-control studies(34,585 cases and 69,544 controls)were conducted along with multipronged experiments to investigate the potential mechanisms by which candidate eRNAQTLs affect CRC risk.Results:A total of 300,112 eRNAQTLs were identified across 30 different cancer types,which exert their influence on eRNA transcription by modulating chromatin status,binding affinity to transcription factors and RNA-binding proteins.These eRNAQTLs were found to be significantly enriched in cancer risk loci,explaining a substantial proportion of cancer heritability.Additionally,tumor-specific eRNAQTLs exhibited high responsiveness to the development of cancer.Moreover,the target genes of these eRNAs were associated with dysregulated signaling pathways and immune cell infiltration in cancer,highlighting their potential as therapeutic targets.Furthermore,multiple ethnic population studies have confirmed that an eRNAQTL rs3094296-T variant decreases the risk of CRC in populations from China(OR=0.91,95%CI 0.88–0.95,P=2.92×10^(-7))and Europe(OR=0.92,95%CI 0.88–0.95,P=4.61×10^(-6)).Mechanistically,rs3094296 had an allele-specific effect on the transcription of the eRNA ENSR00000155786,which functioned as a transcriptional activator promoting the expression of its target gene SENP7.These two genes synergistically suppressed tumor cell proliferation.Our curated list of variants,genes,and drugs has been made available in CancereRNAQTL(http://canernaqtl.whu.edu.cn/#/)to serve as an informative resource for advancing this field.Conclusion:Our findings underscore the significance of eRNAQTLs in transcriptional regulation and disease heritability,pinpointing the potential of eRNA-based therapeutic strategies in cancers.
基金funded by the Deanship of Scientific Research(DSR)at King Abdulaziz University,Jeddah,under grant No.(IFPDP-261-22).
文摘Medical imaging is essential in modern health care,allowing accurate diagnosis and effective treatment planning.These images,however,often demonstrate low contrast,noise,and brightness distortion that reduce their diagnostic reliability.This review presents a structured and comprehensive analysis of advanced histogram equalization(HE)-based techniques for medical image enhancement.Our review methodology encompasses:(1)classical HE approaches and related limitations in medical domains;(2)adaptive schemes like Adaptive Histogram Equalization(AHE)and Contrast Limited Adaptive Histogrma Equalization(CLAHE)and their advance variants;(3)brightnesspreserving schemes like BBHE and MMBEBHE and related algorithms;(4)dynamic and recursive histogram equalization methods incorporating DHE and RMSHE;(5)fuzzy logic-based enhancement methodologies addressing uncertainty and noise in medical images;and(6)hybrid optimization methodologies through the application of metaheuristic algorithms(World Cup Optimization,Particle Swarm Optimization,Genetic Algorithms,along with histogram-based methodologies.)There is also a comparative discussion given based on contrast improvement,image brightness preservation,noise management,and computational efficiency.Such advancements have better capabilities of improving image quality,which is more important for improved diagnosis and image analysis.
文摘Let G=(V,E)be a connected graph.For an integer h≥0,a subset F■V(G)(resp.F■E(G))of G,if any,is called an h-restricted vertex cut(resp.h-restricted edge cut)of G,if G-F is disconnected and every vertex in G-F has at least h neighbors.The cardinality of a minimum h-restricted vertex-cut(resp.h-restricted edge cut)of G is the h-restricted connectivity(resp.h-restricted edge connectivity)of G,and denoted by κ^(h)(G)(resp.λ^(h)(G)).The enhanced hypercube Q_(n,κ)(1≤k≤n)is a variant of the hypercube Q_(n).In this paper,we consider the h-restricted connectivity of Q_(n,κ) for 2≤k≤n-1.Our main results are as follows:(1)κ^(h)(Q_(n,κ))=2^(h)(n-h+1)for 4≤k≤n-1 and 0≤h≤n-3,λ^(h)(Q_(n,κ))=2^(h)(n-h+1)for 2≤k≤n-1 and 0≤h≤n-2.(2)κ^(h)(Q_(n,3))=2^(h-1)(n-h+1)for n≥5 and 4≤h≤n-1,κ^(h)(Q_(n,2))=2^(h-1)(n-h+1)for n≥4 and 3≤h≤n-1.(3)κ^(3)(Q_(n,3))=6n-16 for n≥5,κ^(2)(Q_(n,3))=4n-8 for n≥4 and κ^(2)(Q_(n,2))=3n-5 for n≥3,κ^(1)(Q_(n,3))=2n and κ^(3)(Q_(n,2))=2n-2 for n≥3.
基金funded by the National Natural Science Foundation of China,grant numbers 52374156 and 62476005。
文摘Images taken in dim environments frequently exhibit issues like insufficient brightness,noise,color shifts,and loss of detail.These problems pose significant challenges to dark image enhancement tasks.Current approaches,while effective in global illumination modeling,often struggle to simultaneously suppress noise and preserve structural details,especially under heterogeneous lighting.Furthermore,misalignment between luminance and color channels introduces additional challenges to accurate enhancement.In response to the aforementioned difficulties,we introduce a single-stage framework,M2ATNet,using the multi-scale multi-attention and Transformer architecture.First,to address the problems of texture blurring and residual noise,we design a multi-scale multi-attention denoising module(MMAD),which is applied separately to the luminance and color channels to enhance the structural and texture modeling capabilities.Secondly,to solve the non-alignment problem of the luminance and color channels,we introduce the multi-channel feature fusion Transformer(CFFT)module,which effectively recovers the dark details and corrects the color shifts through cross-channel alignment and deep feature interaction.To guide the model to learn more stably and efficiently,we also fuse multiple types of loss functions to form a hybrid loss term.We extensively evaluate the proposed method on various standard datasets,including LOL-v1,LOL-v2,DICM,LIME,and NPE.Evaluation in terms of numerical metrics and visual quality demonstrate that M2ATNet consistently outperforms existing advanced approaches.Ablation studies further confirm the critical roles played by the MMAD and CFFT modules to detail preservation and visual fidelity under challenging illumination-deficient environments.
文摘Weakly Supervised Semantic Segmentation(WSSS),which relies only on image-level labels,has attracted significant attention for its cost-effectiveness and scalability.Existing methods mainly enhance inter-class distinctions and employ data augmentation to mitigate semantic ambiguity and reduce spurious activations.However,they often neglect the complex contextual dependencies among image patches,resulting in incomplete local representations and limited segmentation accuracy.To address these issues,we propose the Context Patch Fusion with Class Token Enhancement(CPF-CTE)framework,which exploits contextual relations among patches to enrich feature repre-sentations and improve segmentation.At its core,the Contextual-Fusion Bidirectional Long Short-Term Memory(CF-BiLSTM)module captures spatial dependencies between patches and enables bidirectional information flow,yield-ing a more comprehensive understanding of spatial correlations.This strengthens feature learning and segmentation robustness.Moreover,we introduce learnable class tokens that dynamically encode and refine class-specific semantics,enhancing discriminative capability.By effectively integrating spatial and semantic cues,CPF-CTE produces richer and more accurate representations of image content.Extensive experiments on PASCAL VOC 2012 and MS COCO 2014 validate that CPF-CTE consistently surpasses prior WSSS methods.
基金supported by the National Key R&D Program of China(Nos.2023YFC3404200,2023YFC34042012023YFC3404202)+1 种基金the National Natural Science Foundation of China(No.22575253)the Beijing Natural Science Foundation(No.Z220025)。
文摘The development of synthetic hybrid biological systems integrating photosynthetic organisms with organic-abiotic functional materials holds significant promise for enhancing photosynthetic processes.The artificial regulation of the state transition between photosystem I(PSI)and photosystem II(PSII)represents a strategic and promising approach for improving the efficiency of natural photosynthesis.In this study,we demonstrate that poly(benzimidazolium-phenylthiophene)(CP4)featuring a flexible cationic backbone exhibits superior ultraviolet light-harvesting capability.The polymer CP4 enhanced PSI activity in Chlorella pyrenoidosa(C.pyrenoidosa),subsequently promoting PSII activity and augmenting overall photosynthetic performance.During light-dependent reactions,CP4 significantly accelerated photosynthetic electron transfer,resulting in a 330%increase in the production of oxygen and 93%and 96%increases in the ATP and NADPH contents,respectively.In the context of dark reactions,CP4 facilitated the conversion and utilization of light energy,leading to a 6%increase in both carbohydrate and protein contents.These findings indicate that synthetic light-harvesting polymer materials exhibit considerable application potential in the field of biomass production through enhancement of natural photosynthetic efficiency.
基金funded by the National Natural Science Foundation of China(Grant No.62441212)the Major Project of the Natural Science Foundation of Inner Mongolia(Grant No.2025ZD008).
文摘Aiming at the problem of insufficient recognition of implicit variants by existing Chinese sensitive text detection methods,this paper proposes the IPKE-MoE framework,which consists of three parts,namely,a sensitive word variant extraction framework,a sensitive word variant knowledge enhancement layer and a mixture-of-experts(MoE)classification layer.First,sensitive word variants are precisely extracted through dynamic iterative prompt templates and the context-aware capabilities of Large Language Models(LLMs).Next,the extracted variants are used to construct a knowledge enhancement layer for sensitive word variants based on RoCBert models.Specifically,after locating variants via n-gram algorithms,variant types are mapped to embedding vectors and fused with original word vectors.Finally,a mixture-of-experts(MoE)classification layer is designed(sensitive word,sentiment,and semantic experts),which decouples the relationship between sensitiveword existence and text toxicity throughmultiple experts.This framework effectively combines the comprehension ability of Large Language Models(LLMs)with the discriminative ability of smaller models.Our two experiments demonstrate that the sensitive word variant extraction framework based on dynamically iterated prompt templates outperforms other baseline prompt templates.TheRoCBert models incorporating the sensitive word variant knowledge enhancement layer and a mixture-of-experts(MoE)classification layer achieve superior classification performance compared to other baselines.
基金supported in part by the National Natural Science Foundation of China[Grant number 62471075]the Major Science and Technology Project Grant of the Chongqing Municipal Education Commission[Grant number KJZD-M202301901].
文摘Low-light image enhancement aims to improve the visibility of severely degraded images captured under insufficient illumination,alleviating the adverse effects of illumination degradation on image quality.Traditional Retinex-based approaches,inspired by human visual perception of brightness and color,decompose an image into illumination and reflectance components to restore fine details.However,their limited capacity for handling noise and complex lighting conditions often leads to distortions and artifacts in the enhanced results,particularly under extreme low-light scenarios.Although deep learning methods built upon Retinex theory have recently advanced the field,most still suffer frominsufficient interpretability and sub-optimal enhancement performance.This paper presents RetinexWT,a novel framework that tightly integrates classical Retinex theory with modern deep learning.Following Retinex principles,RetinexWT employs wavelet transforms to estimate illumination maps for brightness adjustment.A detail-recovery module that synergistically combines Vision Transformer(ViT)and wavelet transforms is then introduced to guide the restoration of lost details,thereby improving overall image quality.Within the framework,wavelet decomposition splits input features into high-frequency and low-frequency components,enabling scale-specific processing of global illumination/color cues and fine textures.Furthermore,a gating mechanism selectively fuses down-sampled and up-sampled features,while an attention-based fusion strategy enhances model interpretability.Extensive experiments on the LOL dataset demonstrate that RetinexWT surpasses existing Retinex-oriented deeplearning methods,achieving an average Peak Signal-to-Noise Ratio(PSNR)improvement of 0.22 dB over the current StateOfTheArt(SOTA),thereby confirming its superiority in low-light image enhancement.Code is available at https://github.com/CHEN-hJ516/RetinexWT(accessed on 14 October 2025).
基金supported in part by the National Natural Science Foundation of China[Grant number 62471075]the Major Science and Technology Project Grant of the Chongqing Municipal Education Commission[Grant number KJZD-M202301901].
文摘Underwater images often affect the effectiveness of underwater visual tasks due to problems such as light scattering,color distortion,and detail blurring,limiting their application performance.Existing underwater image enhancement methods,although they can improve the image quality to some extent,often lead to problems such as detail loss and edge blurring.To address these problems,we propose FENet,an efficient underwater image enhancement method.FENet first obtains three different scales of images by image downsampling and then transforms them into the frequency domain to extract the low-frequency and high-frequency spectra,respectively.Then,a distance mask and a mean mask are constructed based on the distance and magnitude mean for enhancing the high-frequency part,thus improving the image details and enhancing the effect by suppressing the noise in the low-frequency part.Affected by the light scattering of underwater images and the fact that some details are lost if they are directly reduced to the spatial domain after the frequency domain operation.For this reason,we propose a multi-stage residual feature aggregation module,which focuses on detail extraction and effectively avoids information loss caused by global enhancement.Finally,we combine the edge guidance strategy to further enhance the edge details of the image.Experimental results indicate that FENet outperforms current state-of-the-art underwater image enhancement methods in quantitative and qualitative evaluations on multiple publicly available datasets.
