Background:Locally advanced laryngeal squamous cell carcinoma(LA-LSCC)presents clinical challenges due to the lack of reliable non-invasive biomarkers.This study aimed to evaluate miR-449a as a diagnostic and prognost...Background:Locally advanced laryngeal squamous cell carcinoma(LA-LSCC)presents clinical challenges due to the lack of reliable non-invasive biomarkers.This study aimed to evaluate miR-449a as a diagnostic and prognostic biomarker in LA-LSCC.Methods:miR-449a expression was analyzed in tumor tissues,adjacent normal tissues,and serum from 81 LA-LSCC patients and 50 controls using quantitative real-time reverse transcription polymerase chain reaction(qRT-PCR).We assessed the diagnostic accuracy by Receiver Operating Characteristic curve(ROC curves),clinicopathological associations,survival outcomes(Kaplan-Meier),and treatment response dynamics.Results:miR-449a was significantly downregulated in LA-LSCC tissues(p<0.0001)and serum(p<0.0001),with a strong tissue-serum correlation(R^(2)=0.988).Tissue miR-449a demonstrated a diagnostic accuracy(Area Under the Curve,AUC=0.857),while serum showed moderate accuracy(AUC=0.734).High miR-449a expression correlated with favorable clinicopathological features and improved survival(median overall survival:67.82 vs.23.74 months;p=0.0012).Multivariate analysis confirmed miR-449a as an independent prognostic factor(p<0.001).miR-449a levels increased post-treatment,particularly in responders to chemotherapy/radiation(p<0.0001).Conclusion:miR-449a serves as a non-invasive biomarker for LA-LSCC diagnosis,prognosis,and treatment monitoring.Its dynamic expression highlights potential for risk stratification and therapy response prediction,warranting further validation in larger cohorts.展开更多
With the increasing growth of online news,fake electronic news detection has become one of the most important paradigms of modern research.Traditional electronic news detection techniques are generally based on contex...With the increasing growth of online news,fake electronic news detection has become one of the most important paradigms of modern research.Traditional electronic news detection techniques are generally based on contextual understanding,sequential dependencies,and/or data imbalance.This makes distinction between genuine and fabricated news a challenging task.To address this problem,we propose a novel hybrid architecture,T5-SA-LSTM,which synergistically integrates the T5 Transformer for semantically rich contextual embedding with the Self-Attentionenhanced(SA)Long Short-Term Memory(LSTM).The LSTM is trained using the Adam optimizer,which provides faster and more stable convergence compared to the Stochastic Gradient Descend(SGD)and Root Mean Square Propagation(RMSProp).The WELFake and FakeNewsPrediction datasets are used,which consist of labeled news articles having fake and real news samples.Tokenization and Synthetic Minority Over-sampling Technique(SMOTE)methods are used for data preprocessing to ensure linguistic normalization and class imbalance.The incorporation of the Self-Attention(SA)mechanism enables the model to highlight critical words and phrases,thereby enhancing predictive accuracy.The proposed model is evaluated using accuracy,precision,recall(sensitivity),and F1-score as performance metrics.The model achieved 99%accuracy on the WELFake dataset and 96.5%accuracy on the FakeNewsPrediction dataset.It outperformed the competitive schemes such as T5-SA-LSTM(RMSProp),T5-SA-LSTM(SGD)and some other models.展开更多
CR Dhan 310(CRD310),a biofortified rice variety,contains a significantly higher level of grain protein compared with its recurrent parent Naveen(NV),as well as most adapted high-yielding rice varieties in India.Althou...CR Dhan 310(CRD310),a biofortified rice variety,contains a significantly higher level of grain protein compared with its recurrent parent Naveen(NV),as well as most adapted high-yielding rice varieties in India.Although a limited investigation depicted that CRD310 contained higher levels of glutelin and some essential amino acids,detailed biochemical,molecular,and cellular mechanisms remain to be studied.As one of the means to identify the proteins and understand the underlying mechanism of higher proteins accumulation in grains of CRD310,the comparative proteomics was undertaken on grains of CRD310 and NV at the yellow ripening stage.展开更多
This article is the 19th contribution to the fungal diversity notes series,in which 106 taxa distributed in 3 phyla,11 classes,35 orders,and 64 families are treated.Taxa described in the present study include a new fa...This article is the 19th contribution to the fungal diversity notes series,in which 106 taxa distributed in 3 phyla,11 classes,35 orders,and 64 families are treated.Taxa described in the present study include a new family,5 new genera,69 new species,3 new combinations,25 new host,habitat,and geographical records,a new name,a new collection,as well as reinstating a previously suppressed genus.The newly established family is Parasporidesmiaceae and the five new genera described herein are Dematiodidymosporum,Neoacrogenospora,Parasporidesmium,Speluncomyces,and Uniomyces.The 69 new species are Acrocalymma triseptatum,Agaricus darjeelingensis,Annellophorella aquatica,Anteaglonium menghaiense,Balsamia microspora,Bambusicola dehongensis,Barriopsis menglaense,Benjaminiomyces bergonzoi,Camporesiomyces aquaticus,Camporesiomyces wurfbainiae,Cercospora palmata,Chrysomphalina cantharella,Colletotrichum heteropanacicola,Conioscypha guizhouensis,Conioscypha yadongensis,Cora dalfornoae,Cylindromonium brasiliense,Dematiodidymosporum aquaticum,Distoseptispora dinghuensis,Distoseptispora zunyiensis,Ebollia neocarnea,Eudimeromyces aequatorialis,Eudimeromyces euconni,Funalia indica,Fuscosporella ovalis,Fuscosporella yunnanensis,Halobasidium csapodyae,Halokirschsteiniothelia hunanensis,Hongkongmyces xishuangbannaensis,Inocybe ispartaensis,Laboulbenia neofrancoisiana,Lachnella kunmingensis,Lasmenia thailandica,Leptospora cannabini,Lycoperdon sridharii,Myxospora neomasonii,Natipusilla aquatica,Neoacrogenospora aquatica,Neomassaria sinensis,Neovaginatispora juglandis,Niesslia yunnanensis,Ophiocordyceps aseptatospora,Oxneriaria sheosarensis,Paramicrosphaeropsis vitis,Paramyrothecium strychni,Parapaucispora aquatica,Parasporidesmium aquaticum,Parmelia neosaxatilis,Periconia bambusicola,Periconia neohongheensis,Peroneutypa thailandica,Polyozellus albus,Porina magnoliae,Porostereum subspadiceum,Pseudosperma subvolvatum,Pseudothyridariella caseariae,Rhexocercosporidium ferulae,Russula rubroglutinata,Septoriella iranica,Seriascoma asexuale,Sesquicillium flavum,Sirastachys zhongkaiensis,Speluncomyces lunatus,Sporidesmiella yunnanensis,Striaticonidium xishuangbannaensis,Trametopsis indica,Tulostoma hyderabadensis,Uniomyces hakkeijimanus,and Virgaria guizhouensis.The three new combinations are Lycoperdon alpinum,Lycoperdon lloydii,and Lycoperdon macrogemmae.The 25 new records comprise Acremonium sclerotigenum,Agroathelia rolfsii,Alfaria terrestris,Aspergillus cejpii,Colletotrichum brevisporum,Coriolopsis brunneoleuca,Coriolopsis hainanensis,Cytospora tamaricicola,Fomitopsis malicola,Fulvifomes fastuosus,Fulvifomes thailandicus,Funalia cystidiata,Funalia subgallica,Longididymella vitalbae,Lopharia mirabilis,Metarhizium viridulum,Neopestalotiopsis haikouensis,Occultibambusa aquatica,Phaeoacremonium scolyti,Phaeocytostroma virdimurae,Puccinia mysuruensis,Rhizopus stolonifer,Serpula similis,Trametes ellipsospora,and Vamsapriya shiwandashanensis.In addition,the new name is Irpiciporus pseudoxuchilensis,and the new collection is Aspergillus sydowii.The previously suppressed genus Eudimeromyces has been taxonomically reinstated.展开更多
Origanum elongatum(OE)is an aromatic,medicinal plant endemic to Morocco that is widely used in traditional medicine due to its biological properties.This study aimed to elucidate the chemical composition of the essent...Origanum elongatum(OE)is an aromatic,medicinal plant endemic to Morocco that is widely used in traditional medicine due to its biological properties.This study aimed to elucidate the chemical composition of the essential oil(EO)obtained from O.elongatum(OEEO)at three stages of its life cycle,including vegetative stage(OEEO-VS),flowering stage(OEEO-FS),and post-flowering(OEEO-PFS),as well as to evaluate its biological and antiradical characteristics.The chemical analysis of the essential oil was conducted using gas chromatography-mass spectrometry(GC-MS).The antibacterial activity was evaluated in vitro through distinct methodologies,namely,disc diffusion and microatmosphere assay;subsequently,the minimum inhibitory concentration(MIC)was then determined.The antioxidant potential was also measured by using the DPPH and FRAP assays.The GC-MS revealed the predominant of p-cymene(26.83%_31.45%),γ-terpinene(8.46%_26.95%),thymol(13%_29.54%),and carvacrol(20.25%_37.26%),in all three samples,with notable variations according to the phenological stage of the samples.The EOs extracted at three phenological stages demonstrated notable antibacterial efficacy against all the phytopathogen tested.The MICs for Erwinia amylovora exhibited a range of 6.25 and 250μg/mL.However,for Agrobacterium tumefaciens C58 and Allorhizobium vitis S4,the MICs spanned 125 and 250μg/mL.In the DPPH test,the IC50 values were 168.25±1.14,147.01±0.78,and 132.01±2.06μg/mL for EOs derived from the vegetative,flowering,and post-flowering period,respectively.In the FRAP test,the EC50 values were 164.22±1.04,215.73±1.48,and 184.06±0.95μg/mL for the same stages.The findings offer promising prospects for the phytochemical development,demonstrating how the phenological stage significantly influences the therapeutic and biotechnological potential of O.elongatum.This has the potential to open up new avenues of research in the pharmaceutical,agronomic,and environmental fields.展开更多
Axonal growth inhibitors are released during traumatic injuries to the adult mammalian central nervous system, including after spinal cord injury. These molecules accumulate at the injury site and form a highly inhibi...Axonal growth inhibitors are released during traumatic injuries to the adult mammalian central nervous system, including after spinal cord injury. These molecules accumulate at the injury site and form a highly inhibitory environment for axonal regeneration. Among these inhibitory molecules, myelinassociated inhibitors, including neurite outgrowth inhibitor A, oligodendrocyte myelin glycoprotein, myelin-associated glycoprotein, chondroitin sulfate proteoglycans and repulsive guidance molecule A are of particular importance. Due to their inhibitory nature, they represent exciting molecular targets to study axonal inhibition and regeneration after central injuries. These molecules are mainly produced by neurons, oligodendrocytes, and astrocytes within the scar and in its immediate vicinity. They exert their effects by binding to specific receptors, localized in the membranes of neurons. Receptors for these inhibitory cues include Nogo receptor 1, leucine-rich repeat, and Ig domain containing 1 and p75 neurotrophin receptor/tumor necrosis factor receptor superfamily member 19(that form a receptor complex that binds all myelin-associated inhibitors), and also paired immunoglobulin-like receptor B. Chondroitin sulfate proteoglycans and repulsive guidance molecule A bind to Nogo receptor 1, Nogo receptor 3, receptor protein tyrosine phosphatase σ and leucocyte common antigen related phosphatase, and neogenin, respectively. Once activated, these receptors initiate downstream signaling pathways, the most common amongst them being the Rho A/ROCK signaling pathway. These signaling cascades result in actin depolymerization, neurite outgrowth inhibition, and failure to regenerate after spinal cord injury. Currently, there are no approved pharmacological treatments to overcome spinal cord injuries other than physical rehabilitation and management of the array of symptoms brought on by spinal cord injuries. However, several novel therapies aiming to modulate these inhibitory proteins and/or their receptors are under investigation in ongoing clinical trials. Investigation has also been demonstrating that combinatorial therapies of growth inhibitors with other therapies, such as growth factors or stem-cell therapies, produce stronger results and their potential application in the clinics opens new venues in spinal cord injury treatment.展开更多
Deep neural networks have achieved excellent classification results on several computer vision benchmarks.This has led to the popularity of machine learning as a service,where trained algorithms are hosted on the clou...Deep neural networks have achieved excellent classification results on several computer vision benchmarks.This has led to the popularity of machine learning as a service,where trained algorithms are hosted on the cloud and inference can be obtained on real-world data.In most applications,it is important to compress the vision data due to the enormous bandwidth and memory requirements.Video codecs exploit spatial and temporal correlations to achieve high compression ratios,but they are computationally expensive.This work computes the motion fields between consecutive frames to facilitate the efficient classification of videos.However,contrary to the normal practice of reconstructing the full-resolution frames through motion compensation,this work proposes to infer the class label from the block-based computed motion fields directly.Motion fields are a richer and more complex representation of motion vectors,where each motion vector carries the magnitude and direction information.This approach has two advantages:the cost of motion compensation and video decoding is avoided,and the dimensions of the input signal are highly reduced.This results in a shallower network for classification.The neural network can be trained using motion vectors in two ways:complex representations and magnitude-direction pairs.The proposed work trains a convolutional neural network on the direction and magnitude tensors of the motion fields.Our experimental results show 20×faster convergence during training,reduced overfitting,and accelerated inference on a hand gesture recognition dataset compared to full-resolution and downsampled frames.We validate the proposed methodology on the HGds dataset,achieving a testing accuracy of 99.21%,on the HMDB51 dataset,achieving 82.54%accuracy,and on the UCF101 dataset,achieving 97.13%accuracy,outperforming state-of-the-art methods in computational efficiency.展开更多
It was in the 1980s that research on somatostatin(SST)in Alzheimer’s disease(AD)truly gained traction,demonstrating consistent colocalization with amyloid-β(Aβ),along with massive SST/SST cell losses(Almeida,2024)....It was in the 1980s that research on somatostatin(SST)in Alzheimer’s disease(AD)truly gained traction,demonstrating consistent colocalization with amyloid-β(Aβ),along with massive SST/SST cell losses(Almeida,2024).Although the field already had some grasp over the neuroendocrine and hypothalamic functions of the peptide,very little was known about the GABAergic interneurons(SST-INs)that synthesize it in cortical/hippocampal regions.Quite excitingly,over 40 years later,research has grown effervescent.展开更多
Motor imbalance is a critical failure mode in rotating machinery,potentially causing severe equipment damage if undetected.Traditional vibration-based diagnostic methods rely on direct sensor contact,leading to instal...Motor imbalance is a critical failure mode in rotating machinery,potentially causing severe equipment damage if undetected.Traditional vibration-based diagnostic methods rely on direct sensor contact,leading to installation challenges and measurement artifacts that can compromise accuracy.This study presents a novel radar-based framework for non-contact motor imbalance detection using 24 GHz continuous-wave radar.A dataset of 1802 experimental trials was sourced,covering four imbalance levels(0,10,20,30 g)across varying motor speeds(500–1500 rpm)and load torques(0–3 Nm).Dual-channel in-phase and quadrature radar signals were captured at 10,000 samples per second for 30-s intervals,preserving both amplitude and phase information for analysis.A multi-domain feature extraction methodology captured imbalance signatures in time,frequency,and complex signal domains.From 65 initial features,statistical analysis using Kruskal–Wallis tests identified significant descriptors,and recursive feature elimination with Random Forest reduced the feature set to 20 dimensions,achieving 69%dimensionality reduction without loss of performance.Six machine learning algorithms,Random Forest,Extra Trees Classifier,Extreme Gradient Boosting,Categorical Boosting,Support Vector Machine with radial basis function kernel,and k-Nearest Neighbors were evaluated with grid-search hyperparameter optimization and five-fold cross-validation.The Extra Trees Classifier achieved the best performance with 98.52%test accuracy,98%cross-validation accuracy,and minimal variance,maintaining per-class precision and recall above 97%.Its superior performance is attributed to its randomized split selection and full bootstrapping strategy,which reduce variance and overfitting while effectively capturing the nonlinear feature interactions and non-normal distributions present in the dataset.The model’s average inference time of 70 ms enables near real-time deployment.Comparative analysis demonstrates that the radar-based framework matches or exceeds traditional contact-based methods while eliminating their inherent limitations,providing a robust,scalable,and noninvasive solution for industrial motor condition monitoring,particularly in hazardous or space-constrained environments.展开更多
We present a computer-modeling framework for photovoltaic(PV)source emulation that preserves the exact single-diode physics while enabling iteration-free,real-time evaluation.We derive two closed-form explicit solvers...We present a computer-modeling framework for photovoltaic(PV)source emulation that preserves the exact single-diode physics while enabling iteration-free,real-time evaluation.We derive two closed-form explicit solvers based on the Lambert W function:a voltage-driven V-Lambert solver for high-fidelity I–V computation and a resistance-driven R-Lambert solver designed for seamless integration in a closed-loop PV emulator.Unlike Taylor-linearized explicit models,our proposed formulation retains the exponential nonlinearity of the PV equations.It employs a numerically stable analytical evaluation that eliminates the need for lookup tables and root-finding,all while maintaining limited computational costs and a small memory footprint.The R-Lambert model is integrated into a buck-converter emulator equipped with a discrete PI regulator,which generates current references directly from sensed operating points,thus supporting hardware-constrained implementation.Comprehensive numerical experiments conducted on six commercial modules from various technologies(mono,poly,and multicrystalline)demonstrate significant accuracy improvements under the IEC EN 50530 near-MPP criterion:the V-Lambert solver reduces the±10%Vmpp band error by up to 61 times compared to an explicit-model baseline.Dynamic simulations under varying irradiance,temperature,and load conditions achieve millisecond-scale settling with accurate trajectory tracking.Additionally,processor-in-the-loop experimental validation on an embedded microcontroller supports the simulation results.By unifying exact analytical modeling with embedded realization,this work advances computer modeling for PV emulation,MPPT benchmarking,and controller verification in integrated renewable energy systems.展开更多
Ascorbate(Asc),commonly known as vitamin C,is a vital molecule for plant growth,development,and stress resilience.It is also known to play a crucial role in various physiological processes,including photosynthesis,cel...Ascorbate(Asc),commonly known as vitamin C,is a vital molecule for plant growth,development,and stress resilience.It is also known to play a crucial role in various physiological processes,including photosynthesis,cell division,and differentiation.This article thoroughly explores the processes governing the metabolism of Asc in plants and its roles in physiological functions.It lays down a robust theoretical groundwork for delving into Asc production,transportation,functions,and its potential applications in stress alleviation and horticulture.Furthermore,recent studies indicate that Asc plays a role in regulating fruit development and affecting postharvest storage characteristics,thereby influencing fruit ripening and resilience to stress.Hence,there is a growing importance in studying the synthesis and utilization of Asc in plants.Although the critical role of Asc in controlling plant redox signals has been extensively studied,the precise mechanisms by which it manages cellular redox homeostasis to maintain the equilibrium between reactive oxygen scavenging and cell redox signaling remain elusive.This gap in knowledge presents fresh opportunities to explore how the production of Asc in plants is regulated and how plants react to environmental stressors.Furthermore,this article delves into the potential for a comprehensive investigation into the essential function of Asc in fruits,the development of Asc-rich fruits,and the enhancement of postharvest storage properties.展开更多
Alzheimer’s disease is initially thought to be caused by age-associated accumulation of plaques,in recent years,research has increasingly associated Alzheimer’s disease with lysosomal storage and metabolic disorders...Alzheimer’s disease is initially thought to be caused by age-associated accumulation of plaques,in recent years,research has increasingly associated Alzheimer’s disease with lysosomal storage and metabolic disorders,and the explanation of its pathogenesis has shifted from amyloid and tau accumulation to oxidative stress and impaired lipid and glucose metabolism aggravated by hypoxic conditions.However,the underlying mechanisms linking those cellular processes and conditions to disease progression have yet to be defined.Here,we applied a disease similarity approach to identify unknown molecular targets of Alzheimer’s disease by using transcriptomic data from congenital diseases known to increase Alzheimer’s disease risk,namely Down syndrome,Niemann-Pick type C disease,and mucopolysaccharidoses I.We uncovered common pathways,hub genes,and miRNAs across in vitro and in vivo models of these diseases as potential molecular targets for neuroprotection and amelioration of Alzheimer’s disease pathology,many of which have never been associated with Alzheimer’s disease.We then investigated common molecular alterations in brain samples from a Niemann-Pick type C disease mouse model by juxtaposing them with brain samples of both human and mouse models of Alzheimer’s disease.Detailed phenotypic,molecular,chronological,and biological aging analyses revealed that the Npc1tm(I1061T)Dso mouse model can serve as a potential short-lived in vivo model for brain aging and Alzheimer’s disease research.This research represents the first comprehensive approach to congenital disease association with neurodegeneration and a new perspective on Alzheimer’s disease research while highlighting shortcomings and lack of correlation in diverse in vitro models.Considering the lack of an Alzheimer’s disease mouse model that recapitulates the physiological hallmarks of brain aging,the short-lived Npc1^(tm(I1061T)Dso) mouse model can further accelerate the research in these fields and offer a unique model for understanding the molecular mechanisms of Alzheimer’s disease from a perspective of accelerated brain aging.展开更多
Different forms of programmed cell death have been described to participate in the degeneration of dopaminergic neurons in Parkinson’s disease(PD).Given the critical role that disturbance of mitochondrial homeostasis...Different forms of programmed cell death have been described to participate in the degeneration of dopaminergic neurons in Parkinson’s disease(PD).Given the critical role that disturbance of mitochondrial homeostasis plays in the pathogenesis of PD,apoptosis can be reasonably considered as one of the cell death pathways involved in neuronal loss(Schon and Przedborski,2011).Multiple lines of evidence support that proposal such as the observations in postmortem human brain samples of PD patients including mitochondrial complex I deficiency,reactive oxygen species generation,and oxidative damage to lipids,proteins,and DNA,among others.展开更多
The evolution of cities into digitally managed environments requires computational systems that can operate in real time while supporting predictive and adaptive infrastructure management.Earlier approaches have often...The evolution of cities into digitally managed environments requires computational systems that can operate in real time while supporting predictive and adaptive infrastructure management.Earlier approaches have often advanced one dimension—such as Internet of Things(IoT)-based data acquisition,Artificial Intelligence(AI)-driven analytics,or digital twin visualization—without fully integrating these strands into a single operational loop.As a result,many existing solutions encounter bottlenecks in responsiveness,interoperability,and scalability,while also leaving concerns about data privacy unresolved.This research introduces a hybrid AI–IoT–Digital Twin framework that combines continuous sensing,distributed intelligence,and simulation-based decision support.The design incorporates multi-source sensor data,lightweight edge inference through Convolutional Neural Networks(CNN)and Long ShortTerm Memory(LSTM)models,and federated learning enhanced with secure aggregation and differential privacy to maintain confidentiality.A digital twin layer extends these capabilities by simulating city assets such as traffic flows and water networks,generating what-if scenarios,and issuing actionable control signals.Complementary modules,including model compression and synchronization protocols,are embedded to ensure reliability in bandwidth-constrained and heterogeneous urban environments.The framework is validated in two urban domains:traffic management,where it adapts signal cycles based on real-time congestion patterns,and pipeline monitoring,where it anticipates leaks through pressure and vibration data.Experimental results show a 28%reduction in response time,a 35%decrease in maintenance costs,and a marked reduction in false positives relative to conventional baselines.The architecture also demonstrates stability across 50+edge devices under federated training and resilience to uneven node participation.The proposed system provides a scalable and privacy-aware foundation for predictive urban infrastructure management.By closing the loop between sensing,learning,and control,it reduces operator dependence,enhances resource efficiency,and supports transparent governance models for emerging smart cities.展开更多
The visual system of teleost fish grows continuously,which is a useful model for studying regeneration of the central nervous system.Glial cells are key for this process,but their contribution is still not well define...The visual system of teleost fish grows continuously,which is a useful model for studying regeneration of the central nervous system.Glial cells are key for this process,but their contribution is still not well defined.We followed oligodendrocytes in the visual system of adult zebrafish during regeneration of the optic nerve at 6,24,and 72 hours post-lesion and at 7 and 14 days post-lesion via the sox10:tagRFP transgenic line and confocal microscopy.To understand the changes that these oligodendrocytes undergo during regeneration,we used Sox2 immunohistochemistry,a stem cell marker involved in oligodendrocyte differentiation.We also used the Click-iT™ Plus TUNEL assay to study cell death and a BrdU assay to determine cell proliferation.Before optic nerve crush,sox10:tagRFP oligodendrocytes are located in the retina,in the optic nerve head,and through all the entire optic nerve.Sox2-positive cells are present in the peripheral germinal zone,the mature retina,and the optic nerve.After optic nerve crush,sox10:tagRFP cells disappeared from the optic nerve crush zone,suggesting that they died,although they were not TUNEL positive.Concomitantly,the number of Sox2-positive cells increased around the crushed area,the optic nerve head,and the retina.Then,between 24 hours post-lesion and 14 days post-lesion,double sox10:tagRFP/Sox2-positive cells were detected in the retina,optic nerve head,and whole optic nerve,together with a proliferation response at 72 hours post-lesion.Our results confirm that a degenerating process may occur prior to regeneration.First,sox10:tagRFP oligodendrocytes that surround the degenerated axons stop wrapping them,change their“myelinating oligodendrocyte”morphology to a“nonmyelinating oligodendrocyte”morphology,and die.Then,residual oligodendrocyte progenitor cells in the optic nerve and retina proliferate and differentiate for the purpose of remyelination.As new axons arise from the surviving retinal ganglion cells,new sox10:tagRFP oligodendrocytes arise from residual oligodendrocyte progenitor cells to guide,nourish and myelinate them.Thus,oligodendrocytes play an active role in zebrafish axon regeneration and remyelination.展开更多
Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using d...Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.展开更多
Germinal matrix hemorrhage in preterm neonates often leads to white matter injury,contributing to long-term neurodevelopmental impairments.As resident brain immune cells,microglia play a complex role in injury respons...Germinal matrix hemorrhage in preterm neonates often leads to white matter injury,contributing to long-term neurodevelopmental impairments.As resident brain immune cells,microglia play a complex role in injury response,including inflammation and repair.Although colony-stimulating factor 1 receptor inhibitors such as PLX5622 enable the selective depletion of microglia,their therapeutic potential in neonatal germinal matrix hemorrhage remains underexplored.Here,we used a collagenase-induced germinal matrix hemorrhage model in postnatal day 5 mice,and intraperitoneally administered PLX562272 hours post-germinal matrix hemorrhage to achieve targeted,temporary microglial depletion during the peak injury response.We then assessed the effects of this delayed intervention on oligodendrocyte lineage cell maturation,white matter integrity,and neurobehavioral outcomes.Additionally,RNA sequencing data from a germinal matrix hemorrhage rat model were analyzed using weighted gene co-expression network analysis to identify the critical phases for interventions.RNA sequencing data revealed a critical period in which key synaptic functions declined while immune responses intensified post-germinal matrix hemorrhage,thus pinpointing the critical response phases for potential interventions.Delayed PLX5622 treatment effectively depleted activated microglia,protecting against white matter injury and enhancing oligodendrocyte lineage cell maturation and myelination in subcortical white matter regions.Moreover,magnetic resonance imaging analysis revealed reduced brain lesion volumes in treated mice.Behaviorally,PLX5622-treated mice exhibited significant improvements in motor coordination and reduced hyperactivity compared with vehicle-treated germinal matrix hemorrhage model mice.These findings suggest that,when timed to avoid interference with initial oligodendrocyte lineage cell proliferation,targeted microglial depletion with PLX5622 significantly mitigates white matter damage and improves neurobehavioral outcomes in neonatal germinal matrix hemorrhage.The present study highlights the therapeutic potential of selectively modulating microglial reactivity to support neurodevelopment in preterm infants with brain injury.展开更多
The cloud-fog computing paradigm has emerged as a novel hybrid computing model that integrates computational resources at both fog nodes and cloud servers to address the challenges posed by dynamic and heterogeneous c...The cloud-fog computing paradigm has emerged as a novel hybrid computing model that integrates computational resources at both fog nodes and cloud servers to address the challenges posed by dynamic and heterogeneous computing networks.Finding an optimal computational resource for task offloading and then executing efficiently is a critical issue to achieve a trade-off between energy consumption and transmission delay.In this network,the task processed at fog nodes reduces transmission delay.Still,it increases energy consumption,while routing tasks to the cloud server saves energy at the cost of higher communication delay.Moreover,the order in which offloaded tasks are executed affects the system’s efficiency.For instance,executing lower-priority tasks before higher-priority jobs can disturb the reliability and stability of the system.Therefore,an efficient strategy of optimal computation offloading and task scheduling is required for operational efficacy.In this paper,we introduced a multi-objective and enhanced version of Cheeta Optimizer(CO),namely(MoECO),to jointly optimize the computation offloading and task scheduling in cloud-fog networks to minimize two competing objectives,i.e.,energy consumption and communication delay.MoECO first assigns tasks to the optimal computational nodes and then the allocated tasks are scheduled for processing based on the task priority.The mathematical modelling of CO needs improvement in computation time and convergence speed.Therefore,MoECO is proposed to increase the search capability of agents by controlling the search strategy based on a leader’s location.The adaptive step length operator is adjusted to diversify the solution and thus improves the exploration phase,i.e.,global search strategy.Consequently,this prevents the algorithm from getting trapped in the local optimal solution.Moreover,the interaction factor during the exploitation phase is also adjusted based on the location of the prey instead of the adjacent Cheetah.This increases the exploitation capability of agents,i.e.,local search capability.Furthermore,MoECO employs a multi-objective Pareto-optimal front to simultaneously minimize designated objectives.Comprehensive simulations in MATLAB demonstrate that the proposed algorithm obtains multiple solutions via a Pareto-optimal front and achieves an efficient trade-off between optimization objectives compared to baseline methods.展开更多
Sustainable energy systems will entail a change in the carbon intensity projections,which should be carried out in a proper manner to facilitate the smooth running of the grid and reduce greenhouse emissions.The prese...Sustainable energy systems will entail a change in the carbon intensity projections,which should be carried out in a proper manner to facilitate the smooth running of the grid and reduce greenhouse emissions.The present article outlines the TransCarbonNet,a novel hybrid deep learning framework with self-attention characteristics added to the bidirectional Long Short-Term Memory(Bi-LSTM)network to forecast the carbon intensity of the grid several days.The proposed temporal fusion model not only learns the local temporal interactions but also the long-term patterns of the carbon emission data;hence,it is able to give suitable forecasts over a period of seven days.TransCarbonNet takes advantage of a multi-head self-attention element to identify significant temporal connections,which means the Bi-LSTM element calculates sequential dependencies in both directions.Massive tests on two actual data sets indicate much improved results in comparison with the existing results,with mean relative errors of 15.3 percent and 12.7 percent,respectively.The framework has given explicable weights of attention that reveal critical periods that influence carbon intensity alterations,and informed decisions on the management of carbon sustainability.The effectiveness of the proposed solution has been validated in numerous cases of operations,and TransCarbonNet is established to be an effective tool when it comes to carbon-friendly optimization of the grid.展开更多
Objective:Deep learning is employed increasingly in Gastroenterology(GI)endoscopy computer-aided diagnostics for polyp segmentation and multi-class disease detection.In the real world,implementation requires high accu...Objective:Deep learning is employed increasingly in Gastroenterology(GI)endoscopy computer-aided diagnostics for polyp segmentation and multi-class disease detection.In the real world,implementation requires high accuracy,therapeutically relevant explanations,strong calibration,domain generalization,and efficiency.Current Convolutional Neural Network(CNN)and transformer models compromise border precision and global context,generate attention maps that fail to align with expert reasoning,deteriorate during cross-center changes,and exhibit inadequate calibration,hence diminishing clinical trust.Methods:HMA-DER is a hierarchical multi-attention architecture that uses dilation-enhanced residual blocks and an explainability-aware Cognitive Alignment Score(CAS)regularizer to directly align attribution maps with reasoning signals from experts.The framework has additions that make it more resilient and a way to test for accuracy,macro-averaged F1 score,Area Under the Receiver Operating Characteristic Curve(AUROC),calibration(Expected Calibration Error(ECE),Brier Score),explainability(CAS,insertion/deletion AUC),cross-dataset transfer,and throughput.Results:HMA-DER gets Dice Similarity Coefficient scores of 89.5%and 86.0%on Kvasir-SEG and CVC-ClinicDB,beating the strongest baseline by+1.9 and+1.7 points.It gets 86.4%and 85.3%macro-F1 and 94.0%and 93.4%AUROC on HyperKvasir and GastroVision,which is better than the baseline by+1.4/+1.6macro-F1 and+1.2/+1.1AUROC.Ablation study shows that hierarchical attention gives the highest(+3.0),followed by CAS regularization(+2–3),dilatation(+1.5–2.0),and residual connections(+2–3).Cross-dataset validation demonstrates competitive zero-shot transfer(e.g.,KS→CVC Dice 82.7%),whereas multi-dataset training diminishes the domain gap,yielding an 88.1%primary-metric average.HMA-DER’s mixed-precision inference can handle 155 pictures per second,which helps with calibration.Conclusion:HMA-DER strikes a compromise between accuracy,explainability,robustness,and efficiency for the use of reliable GI computer-aided diagnosis in real-world clinical settings.展开更多
基金The authors extend their appreciation to Taif University,Saudi Arabia,for supporting this work through project No.(TU-DSPP-2024-54).
文摘Background:Locally advanced laryngeal squamous cell carcinoma(LA-LSCC)presents clinical challenges due to the lack of reliable non-invasive biomarkers.This study aimed to evaluate miR-449a as a diagnostic and prognostic biomarker in LA-LSCC.Methods:miR-449a expression was analyzed in tumor tissues,adjacent normal tissues,and serum from 81 LA-LSCC patients and 50 controls using quantitative real-time reverse transcription polymerase chain reaction(qRT-PCR).We assessed the diagnostic accuracy by Receiver Operating Characteristic curve(ROC curves),clinicopathological associations,survival outcomes(Kaplan-Meier),and treatment response dynamics.Results:miR-449a was significantly downregulated in LA-LSCC tissues(p<0.0001)and serum(p<0.0001),with a strong tissue-serum correlation(R^(2)=0.988).Tissue miR-449a demonstrated a diagnostic accuracy(Area Under the Curve,AUC=0.857),while serum showed moderate accuracy(AUC=0.734).High miR-449a expression correlated with favorable clinicopathological features and improved survival(median overall survival:67.82 vs.23.74 months;p=0.0012).Multivariate analysis confirmed miR-449a as an independent prognostic factor(p<0.001).miR-449a levels increased post-treatment,particularly in responders to chemotherapy/radiation(p<0.0001).Conclusion:miR-449a serves as a non-invasive biomarker for LA-LSCC diagnosis,prognosis,and treatment monitoring.Its dynamic expression highlights potential for risk stratification and therapy response prediction,warranting further validation in larger cohorts.
基金supported by Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R195)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘With the increasing growth of online news,fake electronic news detection has become one of the most important paradigms of modern research.Traditional electronic news detection techniques are generally based on contextual understanding,sequential dependencies,and/or data imbalance.This makes distinction between genuine and fabricated news a challenging task.To address this problem,we propose a novel hybrid architecture,T5-SA-LSTM,which synergistically integrates the T5 Transformer for semantically rich contextual embedding with the Self-Attentionenhanced(SA)Long Short-Term Memory(LSTM).The LSTM is trained using the Adam optimizer,which provides faster and more stable convergence compared to the Stochastic Gradient Descend(SGD)and Root Mean Square Propagation(RMSProp).The WELFake and FakeNewsPrediction datasets are used,which consist of labeled news articles having fake and real news samples.Tokenization and Synthetic Minority Over-sampling Technique(SMOTE)methods are used for data preprocessing to ensure linguistic normalization and class imbalance.The incorporation of the Self-Attention(SA)mechanism enables the model to highlight critical words and phrases,thereby enhancing predictive accuracy.The proposed model is evaluated using accuracy,precision,recall(sensitivity),and F1-score as performance metrics.The model achieved 99%accuracy on the WELFake dataset and 96.5%accuracy on the FakeNewsPrediction dataset.It outperformed the competitive schemes such as T5-SA-LSTM(RMSProp),T5-SA-LSTM(SGD)and some other models.
基金supported by the director of Indian Council of Agricultural Research and International Rice Research Institute (ICAR-CRRI), Cuttack, Indiathe coordinator of the ICAR-sponsored project ‘C-reactive protein (CRP) in Biofortification in Selected Crops’, India
文摘CR Dhan 310(CRD310),a biofortified rice variety,contains a significantly higher level of grain protein compared with its recurrent parent Naveen(NV),as well as most adapted high-yielding rice varieties in India.Although a limited investigation depicted that CRD310 contained higher levels of glutelin and some essential amino acids,detailed biochemical,molecular,and cellular mechanisms remain to be studied.As one of the means to identify the proteins and understand the underlying mechanism of higher proteins accumulation in grains of CRD310,the comparative proteomics was undertaken on grains of CRD310 and NV at the yellow ripening stage.
基金the National Natural Science Foundation of China(Grant No.32200015)the Foundation of Guangzhou Municipal Science and Technology Bureau(Grant No.2023A04J1425)+46 种基金the Foundation of Guangzhou Municipal Science and Technology Bureau(Grant No.2023A04J1426)the National Research Council of Thailand(NRCT)Grant“Total fungal diversity in a given forest area with implications towards species numbers,chemical diversity and biotechnology”(Grant No.N42A650547)the Mushroom Research Foundation,Thailand for funding this workthe Distinguished Scientist Fellowship Program(DSFP),King Saud University,Kingdom of Saudi Arabia.Huang Zhang thanks the National Natural Science Foundation of Shandong(Project ID ZR2022MC071 to Huang Zhang)Taishan Scholar Foundation of Shandong Province(tsqn202306276)supported by the Sichuan Science and Technology Program of China(No.2022NSFSC1011)the Scientific and Technical Research Council of Türkiye(TUBİTAK)for the 2219 International Postdoctoral Research Fellowship Programme(Grant No.1059B192202880)the National Science,Research and Innovation Fund:Thailand Science Research Innovation(Basic Research Fund 2021,2022 and 2023)“Biodiversity,taxonomy,phylogeny and evolution of Colletotrichum on avocado,citrus,durian and mango in northern Thailand”,Grant No.652A01003“Biodiversity,taxonomy and phylogeny of Colletotrichum on Citrus and Mango in Northern Thailand”,Grant No.662A01002 and 672A010002the National Natural Science Foundation of China(No.32060012)SERB(CRG/2020/006053),DST,New DelhiInstitution of Eminence(R/Dev./D/IoE/Incentive/2021-22/32387)BHU,Varanasi and Bridge Grant(No.SRICC/Bridge Grant/2024-25/3151),BHU,Varanasi for providing the financial supportsfinancially supported by the“Iranian Mycological Society”the National Natural Science Foundation of China(32260004)the Yunnan Revitalization Talents Support Plan(High-End Foreign Experts Program)Yunnan Provincial Department of Science and Technology“Zhihui Yunnan”Plan(202403AM140023)the Key Laboratory of Yunnan Provincial Department of Education of the Deep-Time Evolution on Biodiversity from the Origin of the Pearl River for their supportthe International Research Support Initiative Program(IRSIP)Schemegrateful to JSPS for an Award of a Postdoctoral Fellowship and the Research Grants No.185701000001 and No.18-06620Extramural Research-SERB,DST(EMR/2016/003078),Government of India for the financial assistancegrateful to‘The PCCF’of Tamil Nadu Forest Department for providing permission(E2/20458/2017),assistance and support during field visit in Eastern Ghats of Tamil NaduRUSA 2.0(Theme-1,Group-1/2021/49)for providing GrantTamil Nadu State Council for Higher Education,Chennai(RGP/2019-20/MU/HECP-0040)for financial assistanceCSIR,New Delhi,India(09/0115(13300)/2022-EMR-I)for the financial assistancethe Beijing Natural Science Foundation-International Scientist Project(Project Number 1S24085)for the financial supportgrateful to DST-PURSE Programme PhaseⅡ,University of Calcutta,India for financial supportChiang Mai University for providing financial support and laboratory facilitiesgrateful to the UP System Balik PhD Program(OVPAA-BPhD2022-02)Grant entitled“Unraveling the hidden diversity of aquatic fungi from Panay Island,Philippines”Govt.of India for financial assistance(BT/PR29521/FCB/125/15/2018)financial support provided by DGAPA-PAPIIT,UNAM(Grant Number IN203524)the Department of Science and Technology,Govt.of India for the Award of the JC Bose Fellowship(Grant No.JCB/2017/000053),DBT-BUILDER(BT/INF/22/SP41176/2020)grant to School of life Sciences,Ministry of EducationGovt.of India and Institution of Excellence Directorate,University of Hyderabad for the award of the Project(Grant No.UOH-IOE-RC3-21-065)and Fellowship(RA)to PVSRN Sarmathe IOE-PDRF(UOH/IOE/SEST/PDRF/1)Grant from University of Hyderabadthe Yunnan Provincial Department of Science and Technology“Zhihui Yunnan”Plan(202403AM140023)the High-Level Talent Recruitment Plan of Yunnan Provinces(High-End Foreign Experts Programs and“Young Talents”)the National Natural Science Foundation of China(No.32460002)the Meemann Chang Academician Workstation in Yunnan Province(202225AF150002)Yunnan Province Young and Middle-aged Academic and Technical Leaders Reserve Talents Program(202305AC350252)Fundacao Arthur Bernardes(FUNARBE)for financial support.the CMU Proactive Researcher,Chiang Mai University(Grant Numbers 796/2567 and EX010059)the Doi Tung Development Project for Sample Collection(Permission Number 7700/17142 with the title‘The diversity of saprobic fungi on selected hosts in forest northern Thailand’),Chiang Rai,ThailandChiang Mai University for partially supportthe support from the Agency of Innovative Development under the Ministry of Higher Education,Science and Innovation of the Republic of Uzbekistan(Project No.AL-8724052922)the National Key R&D Program of China(Project No.2025YFE0104500)The Slovak Grant Agency VEGA(grant No.1/0295/20)for financial supportfinancial support from the Institute of Botany,Jagiellonian University,scientific funds(N18/DBS/000002)financial support by the statutory funds of the W.Szafer Institute of Botany,Polish Academy of Sciences.
文摘This article is the 19th contribution to the fungal diversity notes series,in which 106 taxa distributed in 3 phyla,11 classes,35 orders,and 64 families are treated.Taxa described in the present study include a new family,5 new genera,69 new species,3 new combinations,25 new host,habitat,and geographical records,a new name,a new collection,as well as reinstating a previously suppressed genus.The newly established family is Parasporidesmiaceae and the five new genera described herein are Dematiodidymosporum,Neoacrogenospora,Parasporidesmium,Speluncomyces,and Uniomyces.The 69 new species are Acrocalymma triseptatum,Agaricus darjeelingensis,Annellophorella aquatica,Anteaglonium menghaiense,Balsamia microspora,Bambusicola dehongensis,Barriopsis menglaense,Benjaminiomyces bergonzoi,Camporesiomyces aquaticus,Camporesiomyces wurfbainiae,Cercospora palmata,Chrysomphalina cantharella,Colletotrichum heteropanacicola,Conioscypha guizhouensis,Conioscypha yadongensis,Cora dalfornoae,Cylindromonium brasiliense,Dematiodidymosporum aquaticum,Distoseptispora dinghuensis,Distoseptispora zunyiensis,Ebollia neocarnea,Eudimeromyces aequatorialis,Eudimeromyces euconni,Funalia indica,Fuscosporella ovalis,Fuscosporella yunnanensis,Halobasidium csapodyae,Halokirschsteiniothelia hunanensis,Hongkongmyces xishuangbannaensis,Inocybe ispartaensis,Laboulbenia neofrancoisiana,Lachnella kunmingensis,Lasmenia thailandica,Leptospora cannabini,Lycoperdon sridharii,Myxospora neomasonii,Natipusilla aquatica,Neoacrogenospora aquatica,Neomassaria sinensis,Neovaginatispora juglandis,Niesslia yunnanensis,Ophiocordyceps aseptatospora,Oxneriaria sheosarensis,Paramicrosphaeropsis vitis,Paramyrothecium strychni,Parapaucispora aquatica,Parasporidesmium aquaticum,Parmelia neosaxatilis,Periconia bambusicola,Periconia neohongheensis,Peroneutypa thailandica,Polyozellus albus,Porina magnoliae,Porostereum subspadiceum,Pseudosperma subvolvatum,Pseudothyridariella caseariae,Rhexocercosporidium ferulae,Russula rubroglutinata,Septoriella iranica,Seriascoma asexuale,Sesquicillium flavum,Sirastachys zhongkaiensis,Speluncomyces lunatus,Sporidesmiella yunnanensis,Striaticonidium xishuangbannaensis,Trametopsis indica,Tulostoma hyderabadensis,Uniomyces hakkeijimanus,and Virgaria guizhouensis.The three new combinations are Lycoperdon alpinum,Lycoperdon lloydii,and Lycoperdon macrogemmae.The 25 new records comprise Acremonium sclerotigenum,Agroathelia rolfsii,Alfaria terrestris,Aspergillus cejpii,Colletotrichum brevisporum,Coriolopsis brunneoleuca,Coriolopsis hainanensis,Cytospora tamaricicola,Fomitopsis malicola,Fulvifomes fastuosus,Fulvifomes thailandicus,Funalia cystidiata,Funalia subgallica,Longididymella vitalbae,Lopharia mirabilis,Metarhizium viridulum,Neopestalotiopsis haikouensis,Occultibambusa aquatica,Phaeoacremonium scolyti,Phaeocytostroma virdimurae,Puccinia mysuruensis,Rhizopus stolonifer,Serpula similis,Trametes ellipsospora,and Vamsapriya shiwandashanensis.In addition,the new name is Irpiciporus pseudoxuchilensis,and the new collection is Aspergillus sydowii.The previously suppressed genus Eudimeromyces has been taxonomically reinstated.
文摘Origanum elongatum(OE)is an aromatic,medicinal plant endemic to Morocco that is widely used in traditional medicine due to its biological properties.This study aimed to elucidate the chemical composition of the essential oil(EO)obtained from O.elongatum(OEEO)at three stages of its life cycle,including vegetative stage(OEEO-VS),flowering stage(OEEO-FS),and post-flowering(OEEO-PFS),as well as to evaluate its biological and antiradical characteristics.The chemical analysis of the essential oil was conducted using gas chromatography-mass spectrometry(GC-MS).The antibacterial activity was evaluated in vitro through distinct methodologies,namely,disc diffusion and microatmosphere assay;subsequently,the minimum inhibitory concentration(MIC)was then determined.The antioxidant potential was also measured by using the DPPH and FRAP assays.The GC-MS revealed the predominant of p-cymene(26.83%_31.45%),γ-terpinene(8.46%_26.95%),thymol(13%_29.54%),and carvacrol(20.25%_37.26%),in all three samples,with notable variations according to the phenological stage of the samples.The EOs extracted at three phenological stages demonstrated notable antibacterial efficacy against all the phytopathogen tested.The MICs for Erwinia amylovora exhibited a range of 6.25 and 250μg/mL.However,for Agrobacterium tumefaciens C58 and Allorhizobium vitis S4,the MICs spanned 125 and 250μg/mL.In the DPPH test,the IC50 values were 168.25±1.14,147.01±0.78,and 132.01±2.06μg/mL for EOs derived from the vegetative,flowering,and post-flowering period,respectively.In the FRAP test,the EC50 values were 164.22±1.04,215.73±1.48,and 184.06±0.95μg/mL for the same stages.The findings offer promising prospects for the phytochemical development,demonstrating how the phenological stage significantly influences the therapeutic and biotechnological potential of O.elongatum.This has the potential to open up new avenues of research in the pharmaceutical,agronomic,and environmental fields.
基金a Ph D fellowship by FCT-Fundacao para a Ciência Tecnologia (SFRH/BD/135868/2018)(to SSC)。
文摘Axonal growth inhibitors are released during traumatic injuries to the adult mammalian central nervous system, including after spinal cord injury. These molecules accumulate at the injury site and form a highly inhibitory environment for axonal regeneration. Among these inhibitory molecules, myelinassociated inhibitors, including neurite outgrowth inhibitor A, oligodendrocyte myelin glycoprotein, myelin-associated glycoprotein, chondroitin sulfate proteoglycans and repulsive guidance molecule A are of particular importance. Due to their inhibitory nature, they represent exciting molecular targets to study axonal inhibition and regeneration after central injuries. These molecules are mainly produced by neurons, oligodendrocytes, and astrocytes within the scar and in its immediate vicinity. They exert their effects by binding to specific receptors, localized in the membranes of neurons. Receptors for these inhibitory cues include Nogo receptor 1, leucine-rich repeat, and Ig domain containing 1 and p75 neurotrophin receptor/tumor necrosis factor receptor superfamily member 19(that form a receptor complex that binds all myelin-associated inhibitors), and also paired immunoglobulin-like receptor B. Chondroitin sulfate proteoglycans and repulsive guidance molecule A bind to Nogo receptor 1, Nogo receptor 3, receptor protein tyrosine phosphatase σ and leucocyte common antigen related phosphatase, and neogenin, respectively. Once activated, these receptors initiate downstream signaling pathways, the most common amongst them being the Rho A/ROCK signaling pathway. These signaling cascades result in actin depolymerization, neurite outgrowth inhibition, and failure to regenerate after spinal cord injury. Currently, there are no approved pharmacological treatments to overcome spinal cord injuries other than physical rehabilitation and management of the array of symptoms brought on by spinal cord injuries. However, several novel therapies aiming to modulate these inhibitory proteins and/or their receptors are under investigation in ongoing clinical trials. Investigation has also been demonstrating that combinatorial therapies of growth inhibitors with other therapies, such as growth factors or stem-cell therapies, produce stronger results and their potential application in the clinics opens new venues in spinal cord injury treatment.
基金Supported by Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R896).
文摘Deep neural networks have achieved excellent classification results on several computer vision benchmarks.This has led to the popularity of machine learning as a service,where trained algorithms are hosted on the cloud and inference can be obtained on real-world data.In most applications,it is important to compress the vision data due to the enormous bandwidth and memory requirements.Video codecs exploit spatial and temporal correlations to achieve high compression ratios,but they are computationally expensive.This work computes the motion fields between consecutive frames to facilitate the efficient classification of videos.However,contrary to the normal practice of reconstructing the full-resolution frames through motion compensation,this work proposes to infer the class label from the block-based computed motion fields directly.Motion fields are a richer and more complex representation of motion vectors,where each motion vector carries the magnitude and direction information.This approach has two advantages:the cost of motion compensation and video decoding is avoided,and the dimensions of the input signal are highly reduced.This results in a shallower network for classification.The neural network can be trained using motion vectors in two ways:complex representations and magnitude-direction pairs.The proposed work trains a convolutional neural network on the direction and magnitude tensors of the motion fields.Our experimental results show 20×faster convergence during training,reduced overfitting,and accelerated inference on a hand gesture recognition dataset compared to full-resolution and downsampled frames.We validate the proposed methodology on the HGds dataset,achieving a testing accuracy of 99.21%,on the HMDB51 dataset,achieving 82.54%accuracy,and on the UCF101 dataset,achieving 97.13%accuracy,outperforming state-of-the-art methods in computational efficiency.
文摘It was in the 1980s that research on somatostatin(SST)in Alzheimer’s disease(AD)truly gained traction,demonstrating consistent colocalization with amyloid-β(Aβ),along with massive SST/SST cell losses(Almeida,2024).Although the field already had some grasp over the neuroendocrine and hypothalamic functions of the peptide,very little was known about the GABAergic interneurons(SST-INs)that synthesize it in cortical/hippocampal regions.Quite excitingly,over 40 years later,research has grown effervescent.
基金funded by Princess Nourah bint Abdulrahman University Researchers Support-ing Project number(PNURSP2026R346)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Motor imbalance is a critical failure mode in rotating machinery,potentially causing severe equipment damage if undetected.Traditional vibration-based diagnostic methods rely on direct sensor contact,leading to installation challenges and measurement artifacts that can compromise accuracy.This study presents a novel radar-based framework for non-contact motor imbalance detection using 24 GHz continuous-wave radar.A dataset of 1802 experimental trials was sourced,covering four imbalance levels(0,10,20,30 g)across varying motor speeds(500–1500 rpm)and load torques(0–3 Nm).Dual-channel in-phase and quadrature radar signals were captured at 10,000 samples per second for 30-s intervals,preserving both amplitude and phase information for analysis.A multi-domain feature extraction methodology captured imbalance signatures in time,frequency,and complex signal domains.From 65 initial features,statistical analysis using Kruskal–Wallis tests identified significant descriptors,and recursive feature elimination with Random Forest reduced the feature set to 20 dimensions,achieving 69%dimensionality reduction without loss of performance.Six machine learning algorithms,Random Forest,Extra Trees Classifier,Extreme Gradient Boosting,Categorical Boosting,Support Vector Machine with radial basis function kernel,and k-Nearest Neighbors were evaluated with grid-search hyperparameter optimization and five-fold cross-validation.The Extra Trees Classifier achieved the best performance with 98.52%test accuracy,98%cross-validation accuracy,and minimal variance,maintaining per-class precision and recall above 97%.Its superior performance is attributed to its randomized split selection and full bootstrapping strategy,which reduce variance and overfitting while effectively capturing the nonlinear feature interactions and non-normal distributions present in the dataset.The model’s average inference time of 70 ms enables near real-time deployment.Comparative analysis demonstrates that the radar-based framework matches or exceeds traditional contact-based methods while eliminating their inherent limitations,providing a robust,scalable,and noninvasive solution for industrial motor condition monitoring,particularly in hazardous or space-constrained environments.
基金funded by Scientific Research Deanship at University of Ha’il-Saudi Arabia through project number(RG-24014).
文摘We present a computer-modeling framework for photovoltaic(PV)source emulation that preserves the exact single-diode physics while enabling iteration-free,real-time evaluation.We derive two closed-form explicit solvers based on the Lambert W function:a voltage-driven V-Lambert solver for high-fidelity I–V computation and a resistance-driven R-Lambert solver designed for seamless integration in a closed-loop PV emulator.Unlike Taylor-linearized explicit models,our proposed formulation retains the exponential nonlinearity of the PV equations.It employs a numerically stable analytical evaluation that eliminates the need for lookup tables and root-finding,all while maintaining limited computational costs and a small memory footprint.The R-Lambert model is integrated into a buck-converter emulator equipped with a discrete PI regulator,which generates current references directly from sensed operating points,thus supporting hardware-constrained implementation.Comprehensive numerical experiments conducted on six commercial modules from various technologies(mono,poly,and multicrystalline)demonstrate significant accuracy improvements under the IEC EN 50530 near-MPP criterion:the V-Lambert solver reduces the±10%Vmpp band error by up to 61 times compared to an explicit-model baseline.Dynamic simulations under varying irradiance,temperature,and load conditions achieve millisecond-scale settling with accurate trajectory tracking.Additionally,processor-in-the-loop experimental validation on an embedded microcontroller supports the simulation results.By unifying exact analytical modeling with embedded realization,this work advances computer modeling for PV emulation,MPPT benchmarking,and controller verification in integrated renewable energy systems.
基金supported by the Lendület/Momentum Programme of the Hungarian Academy of Sciencesthe National Research, Development, and Innovation Office, Hungary (Grant Nos. LP2024/21 and K146791)+2 种基金Bayers fellowship program MEDHA and Department of Botany, University of Calicutthe financial assistance provided in the form of Junior Research Fellowship from the University Grants Commission (UGC), Indiathe financial assistance provided by the Council for Scientific and Industrial Research(CSIR), India
文摘Ascorbate(Asc),commonly known as vitamin C,is a vital molecule for plant growth,development,and stress resilience.It is also known to play a crucial role in various physiological processes,including photosynthesis,cell division,and differentiation.This article thoroughly explores the processes governing the metabolism of Asc in plants and its roles in physiological functions.It lays down a robust theoretical groundwork for delving into Asc production,transportation,functions,and its potential applications in stress alleviation and horticulture.Furthermore,recent studies indicate that Asc plays a role in regulating fruit development and affecting postharvest storage characteristics,thereby influencing fruit ripening and resilience to stress.Hence,there is a growing importance in studying the synthesis and utilization of Asc in plants.Although the critical role of Asc in controlling plant redox signals has been extensively studied,the precise mechanisms by which it manages cellular redox homeostasis to maintain the equilibrium between reactive oxygen scavenging and cell redox signaling remain elusive.This gap in knowledge presents fresh opportunities to explore how the production of Asc in plants is regulated and how plants react to environmental stressors.Furthermore,this article delves into the potential for a comprehensive investigation into the essential function of Asc in fruits,the development of Asc-rich fruits,and the enhancement of postharvest storage properties.
基金supported by the NIA/NIH(1K01AG060040).Studies performed by JN were funded by the NICHD/NIH(5R00HD096117)Microscopy Core Facility supported,in part,with funding from NIH-NCI Cancer Center Support Grant P30 CA016059.
文摘Alzheimer’s disease is initially thought to be caused by age-associated accumulation of plaques,in recent years,research has increasingly associated Alzheimer’s disease with lysosomal storage and metabolic disorders,and the explanation of its pathogenesis has shifted from amyloid and tau accumulation to oxidative stress and impaired lipid and glucose metabolism aggravated by hypoxic conditions.However,the underlying mechanisms linking those cellular processes and conditions to disease progression have yet to be defined.Here,we applied a disease similarity approach to identify unknown molecular targets of Alzheimer’s disease by using transcriptomic data from congenital diseases known to increase Alzheimer’s disease risk,namely Down syndrome,Niemann-Pick type C disease,and mucopolysaccharidoses I.We uncovered common pathways,hub genes,and miRNAs across in vitro and in vivo models of these diseases as potential molecular targets for neuroprotection and amelioration of Alzheimer’s disease pathology,many of which have never been associated with Alzheimer’s disease.We then investigated common molecular alterations in brain samples from a Niemann-Pick type C disease mouse model by juxtaposing them with brain samples of both human and mouse models of Alzheimer’s disease.Detailed phenotypic,molecular,chronological,and biological aging analyses revealed that the Npc1tm(I1061T)Dso mouse model can serve as a potential short-lived in vivo model for brain aging and Alzheimer’s disease research.This research represents the first comprehensive approach to congenital disease association with neurodegeneration and a new perspective on Alzheimer’s disease research while highlighting shortcomings and lack of correlation in diverse in vitro models.Considering the lack of an Alzheimer’s disease mouse model that recapitulates the physiological hallmarks of brain aging,the short-lived Npc1^(tm(I1061T)Dso) mouse model can further accelerate the research in these fields and offer a unique model for understanding the molecular mechanisms of Alzheimer’s disease from a perspective of accelerated brain aging.
基金supported by the Spanish Ministerio de Ciencia e Innovación/Agencia Española de Investigación(PID2021-124096OB-I00)(to JLV)JGR was granted by Demensfonden,The Royal Physiografic Society of Lund,Neurofonden,The Greta och Johan Kocks stiftelser,and Bertil och Ebon Norlins stiftelse.
文摘Different forms of programmed cell death have been described to participate in the degeneration of dopaminergic neurons in Parkinson’s disease(PD).Given the critical role that disturbance of mitochondrial homeostasis plays in the pathogenesis of PD,apoptosis can be reasonably considered as one of the cell death pathways involved in neuronal loss(Schon and Przedborski,2011).Multiple lines of evidence support that proposal such as the observations in postmortem human brain samples of PD patients including mitochondrial complex I deficiency,reactive oxygen species generation,and oxidative damage to lipids,proteins,and DNA,among others.
基金The researchers would like to thank the Deanship of Graduate Studies and Scientific Research at Qassim University for financial support(QU-APC-2025)。
文摘The evolution of cities into digitally managed environments requires computational systems that can operate in real time while supporting predictive and adaptive infrastructure management.Earlier approaches have often advanced one dimension—such as Internet of Things(IoT)-based data acquisition,Artificial Intelligence(AI)-driven analytics,or digital twin visualization—without fully integrating these strands into a single operational loop.As a result,many existing solutions encounter bottlenecks in responsiveness,interoperability,and scalability,while also leaving concerns about data privacy unresolved.This research introduces a hybrid AI–IoT–Digital Twin framework that combines continuous sensing,distributed intelligence,and simulation-based decision support.The design incorporates multi-source sensor data,lightweight edge inference through Convolutional Neural Networks(CNN)and Long ShortTerm Memory(LSTM)models,and federated learning enhanced with secure aggregation and differential privacy to maintain confidentiality.A digital twin layer extends these capabilities by simulating city assets such as traffic flows and water networks,generating what-if scenarios,and issuing actionable control signals.Complementary modules,including model compression and synchronization protocols,are embedded to ensure reliability in bandwidth-constrained and heterogeneous urban environments.The framework is validated in two urban domains:traffic management,where it adapts signal cycles based on real-time congestion patterns,and pipeline monitoring,where it anticipates leaks through pressure and vibration data.Experimental results show a 28%reduction in response time,a 35%decrease in maintenance costs,and a marked reduction in false positives relative to conventional baselines.The architecture also demonstrates stability across 50+edge devices under federated training and resilience to uneven node participation.The proposed system provides a scalable and privacy-aware foundation for predictive urban infrastructure management.By closing the loop between sensing,learning,and control,it reduces operator dependence,enhances resource efficiency,and supports transparent governance models for emerging smart cities.
基金supported by the Lanzadera TCUE and C2 program(Universidad de Salamanca)(to ASL)the Spanish National Research Council(CSIC)funded by the Junta de Castilla y León and co-financed by the European Regional Development Fund(ERDF“Europe drives our growth”):Internationalization Project“CL-EI-2021-08-IBFG Unit of Excellence”,Grant(PID2022-138478OA-100)funded by MICIU/AEI/10.13039/501100011033 and,by FEDER,UE(to MGM)+3 种基金Junta de Castilla y León(SA225P23)Gerencia Regional de Salud(2701/A1/2023)(to AV)the Plan Especial Grado Medicina(USAL)(to CPM)a Ramón y Cajal researcher:Grant RYC2021-033684-I funded by MICIU/AEI/10.13039/501100011033 and,by European Union NextGenerationEU/PRTR.
文摘The visual system of teleost fish grows continuously,which is a useful model for studying regeneration of the central nervous system.Glial cells are key for this process,but their contribution is still not well defined.We followed oligodendrocytes in the visual system of adult zebrafish during regeneration of the optic nerve at 6,24,and 72 hours post-lesion and at 7 and 14 days post-lesion via the sox10:tagRFP transgenic line and confocal microscopy.To understand the changes that these oligodendrocytes undergo during regeneration,we used Sox2 immunohistochemistry,a stem cell marker involved in oligodendrocyte differentiation.We also used the Click-iT™ Plus TUNEL assay to study cell death and a BrdU assay to determine cell proliferation.Before optic nerve crush,sox10:tagRFP oligodendrocytes are located in the retina,in the optic nerve head,and through all the entire optic nerve.Sox2-positive cells are present in the peripheral germinal zone,the mature retina,and the optic nerve.After optic nerve crush,sox10:tagRFP cells disappeared from the optic nerve crush zone,suggesting that they died,although they were not TUNEL positive.Concomitantly,the number of Sox2-positive cells increased around the crushed area,the optic nerve head,and the retina.Then,between 24 hours post-lesion and 14 days post-lesion,double sox10:tagRFP/Sox2-positive cells were detected in the retina,optic nerve head,and whole optic nerve,together with a proliferation response at 72 hours post-lesion.Our results confirm that a degenerating process may occur prior to regeneration.First,sox10:tagRFP oligodendrocytes that surround the degenerated axons stop wrapping them,change their“myelinating oligodendrocyte”morphology to a“nonmyelinating oligodendrocyte”morphology,and die.Then,residual oligodendrocyte progenitor cells in the optic nerve and retina proliferate and differentiate for the purpose of remyelination.As new axons arise from the surviving retinal ganglion cells,new sox10:tagRFP oligodendrocytes arise from residual oligodendrocyte progenitor cells to guide,nourish and myelinate them.Thus,oligodendrocytes play an active role in zebrafish axon regeneration and remyelination.
基金The work described in this paper was fully supported by a grant from Hong Kong Metropolitan University(RIF/2021/05).
文摘Parkinson’s disease(PD)is a debilitating neurological disorder affecting over 10 million people worldwide.PD classification models using voice signals as input are common in the literature.It is believed that using deep learning algorithms further enhances performance;nevertheless,it is challenging due to the nature of small-scale and imbalanced PD datasets.This paper proposed a convolutional neural network-based deep support vector machine(CNN-DSVM)to automate the feature extraction process using CNN and extend the conventional SVM to a DSVM for better classification performance in small-scale PD datasets.A customized kernel function reduces the impact of biased classification towards the majority class(healthy candidates in our consideration).An improved generative adversarial network(IGAN)was designed to generate additional training data to enhance the model’s performance.For performance evaluation,the proposed algorithm achieves a sensitivity of 97.6%and a specificity of 97.3%.The performance comparison is evaluated from five perspectives,including comparisons with different data generation algorithms,feature extraction techniques,kernel functions,and existing works.Results reveal the effectiveness of the IGAN algorithm,which improves the sensitivity and specificity by 4.05%–4.72%and 4.96%–5.86%,respectively;and the effectiveness of the CNN-DSVM algorithm,which improves the sensitivity by 1.24%–57.4%and specificity by 1.04%–163%and reduces biased detection towards the majority class.The ablation experiments confirm the effectiveness of individual components.Two future research directions have also been suggested.
基金supported by the National Key Research and Development Program of China,No.2022YFC2704801(to CZhu)the National Natural Science Foundation of China,Nos.U21A20347(to CZhu),82203969(to YX),82371472(to XZ)+3 种基金Health Commission of Henan Province,Nos.SBGJ202303039(to XZ),SBGJ202301009(to CZhu),YQRC2024018(to XZ),YQRC2024019(to YX)Henan Science and Technology Department,Nos.242102311054(to XZ),241111521300(to CZhu),GZS2023003(to XW)Swedish Research Council,Nos.2022-01019(to CZhu),2021-01950(to XW)Swedish Governmental Grants to Scientists Working in Healthcare,Nos.ALFGBG-1005209(to CZhu),ALFBG-1005257(to XW),ALFGBG-965197(to CZhu).
文摘Germinal matrix hemorrhage in preterm neonates often leads to white matter injury,contributing to long-term neurodevelopmental impairments.As resident brain immune cells,microglia play a complex role in injury response,including inflammation and repair.Although colony-stimulating factor 1 receptor inhibitors such as PLX5622 enable the selective depletion of microglia,their therapeutic potential in neonatal germinal matrix hemorrhage remains underexplored.Here,we used a collagenase-induced germinal matrix hemorrhage model in postnatal day 5 mice,and intraperitoneally administered PLX562272 hours post-germinal matrix hemorrhage to achieve targeted,temporary microglial depletion during the peak injury response.We then assessed the effects of this delayed intervention on oligodendrocyte lineage cell maturation,white matter integrity,and neurobehavioral outcomes.Additionally,RNA sequencing data from a germinal matrix hemorrhage rat model were analyzed using weighted gene co-expression network analysis to identify the critical phases for interventions.RNA sequencing data revealed a critical period in which key synaptic functions declined while immune responses intensified post-germinal matrix hemorrhage,thus pinpointing the critical response phases for potential interventions.Delayed PLX5622 treatment effectively depleted activated microglia,protecting against white matter injury and enhancing oligodendrocyte lineage cell maturation and myelination in subcortical white matter regions.Moreover,magnetic resonance imaging analysis revealed reduced brain lesion volumes in treated mice.Behaviorally,PLX5622-treated mice exhibited significant improvements in motor coordination and reduced hyperactivity compared with vehicle-treated germinal matrix hemorrhage model mice.These findings suggest that,when timed to avoid interference with initial oligodendrocyte lineage cell proliferation,targeted microglial depletion with PLX5622 significantly mitigates white matter damage and improves neurobehavioral outcomes in neonatal germinal matrix hemorrhage.The present study highlights the therapeutic potential of selectively modulating microglial reactivity to support neurodevelopment in preterm infants with brain injury.
基金appreciation to the Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R384)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘The cloud-fog computing paradigm has emerged as a novel hybrid computing model that integrates computational resources at both fog nodes and cloud servers to address the challenges posed by dynamic and heterogeneous computing networks.Finding an optimal computational resource for task offloading and then executing efficiently is a critical issue to achieve a trade-off between energy consumption and transmission delay.In this network,the task processed at fog nodes reduces transmission delay.Still,it increases energy consumption,while routing tasks to the cloud server saves energy at the cost of higher communication delay.Moreover,the order in which offloaded tasks are executed affects the system’s efficiency.For instance,executing lower-priority tasks before higher-priority jobs can disturb the reliability and stability of the system.Therefore,an efficient strategy of optimal computation offloading and task scheduling is required for operational efficacy.In this paper,we introduced a multi-objective and enhanced version of Cheeta Optimizer(CO),namely(MoECO),to jointly optimize the computation offloading and task scheduling in cloud-fog networks to minimize two competing objectives,i.e.,energy consumption and communication delay.MoECO first assigns tasks to the optimal computational nodes and then the allocated tasks are scheduled for processing based on the task priority.The mathematical modelling of CO needs improvement in computation time and convergence speed.Therefore,MoECO is proposed to increase the search capability of agents by controlling the search strategy based on a leader’s location.The adaptive step length operator is adjusted to diversify the solution and thus improves the exploration phase,i.e.,global search strategy.Consequently,this prevents the algorithm from getting trapped in the local optimal solution.Moreover,the interaction factor during the exploitation phase is also adjusted based on the location of the prey instead of the adjacent Cheetah.This increases the exploitation capability of agents,i.e.,local search capability.Furthermore,MoECO employs a multi-objective Pareto-optimal front to simultaneously minimize designated objectives.Comprehensive simulations in MATLAB demonstrate that the proposed algorithm obtains multiple solutions via a Pareto-optimal front and achieves an efficient trade-off between optimization objectives compared to baseline methods.
基金funded by the Deanship of Scientific Research and Libraries at Princess Nourah bint Abdulrahman University,through the“Nafea”Program,Grant No.(NP-45-082).
文摘Sustainable energy systems will entail a change in the carbon intensity projections,which should be carried out in a proper manner to facilitate the smooth running of the grid and reduce greenhouse emissions.The present article outlines the TransCarbonNet,a novel hybrid deep learning framework with self-attention characteristics added to the bidirectional Long Short-Term Memory(Bi-LSTM)network to forecast the carbon intensity of the grid several days.The proposed temporal fusion model not only learns the local temporal interactions but also the long-term patterns of the carbon emission data;hence,it is able to give suitable forecasts over a period of seven days.TransCarbonNet takes advantage of a multi-head self-attention element to identify significant temporal connections,which means the Bi-LSTM element calculates sequential dependencies in both directions.Massive tests on two actual data sets indicate much improved results in comparison with the existing results,with mean relative errors of 15.3 percent and 12.7 percent,respectively.The framework has given explicable weights of attention that reveal critical periods that influence carbon intensity alterations,and informed decisions on the management of carbon sustainability.The effectiveness of the proposed solution has been validated in numerous cases of operations,and TransCarbonNet is established to be an effective tool when it comes to carbon-friendly optimization of the grid.
文摘Objective:Deep learning is employed increasingly in Gastroenterology(GI)endoscopy computer-aided diagnostics for polyp segmentation and multi-class disease detection.In the real world,implementation requires high accuracy,therapeutically relevant explanations,strong calibration,domain generalization,and efficiency.Current Convolutional Neural Network(CNN)and transformer models compromise border precision and global context,generate attention maps that fail to align with expert reasoning,deteriorate during cross-center changes,and exhibit inadequate calibration,hence diminishing clinical trust.Methods:HMA-DER is a hierarchical multi-attention architecture that uses dilation-enhanced residual blocks and an explainability-aware Cognitive Alignment Score(CAS)regularizer to directly align attribution maps with reasoning signals from experts.The framework has additions that make it more resilient and a way to test for accuracy,macro-averaged F1 score,Area Under the Receiver Operating Characteristic Curve(AUROC),calibration(Expected Calibration Error(ECE),Brier Score),explainability(CAS,insertion/deletion AUC),cross-dataset transfer,and throughput.Results:HMA-DER gets Dice Similarity Coefficient scores of 89.5%and 86.0%on Kvasir-SEG and CVC-ClinicDB,beating the strongest baseline by+1.9 and+1.7 points.It gets 86.4%and 85.3%macro-F1 and 94.0%and 93.4%AUROC on HyperKvasir and GastroVision,which is better than the baseline by+1.4/+1.6macro-F1 and+1.2/+1.1AUROC.Ablation study shows that hierarchical attention gives the highest(+3.0),followed by CAS regularization(+2–3),dilatation(+1.5–2.0),and residual connections(+2–3).Cross-dataset validation demonstrates competitive zero-shot transfer(e.g.,KS→CVC Dice 82.7%),whereas multi-dataset training diminishes the domain gap,yielding an 88.1%primary-metric average.HMA-DER’s mixed-precision inference can handle 155 pictures per second,which helps with calibration.Conclusion:HMA-DER strikes a compromise between accuracy,explainability,robustness,and efficiency for the use of reliable GI computer-aided diagnosis in real-world clinical settings.
