The study was conducted to identify indoor air quality and the level of thermal comfort in various selected locations in Faculty of Engineering and Built Environment （FKAB）, University Kebangsaan Malaysia （UKM） wi...The study was conducted to identify indoor air quality and the level of thermal comfort in various selected locations in Faculty of Engineering and Built Environment （FKAB）, University Kebangsaan Malaysia （UKM） with built-up area of 250,936 fie. The indoor air quality and thermal comfort were measured at various selected locations by using indoor air quality equipment （Thermal Comfort SERI）. The thermal comfort assessments are based on Malaysian Code of Practice Indoor Air Quality 2005 and Moderate Thermal Environments-Determination of the PMV and PPD indices specification of the condition for thermal comfort （ISO7730：1994） From the data analysis, the FKAB building is considered inadequately vented space. The concentration of CO2 for all sampling area evaluated exceeds the recommended concentration （〉 1000 ppm）. The ventilation system used in FKAB building is designed by delivering fix amount of fresh air into building from external building without consideration on the number of occupants. This common ventilation design will increase the amount of CO2 dramatically all day long and these reflect the inefficiency of energy used. The faculty needs to be equipped with a comprehensive energy management system that can allow detailed documentation of continuous performance of all energy system and consumption in the building.展开更多
The notoriety of the shortage of qualified professionals in the engineering segment to meet the existing projects and also the future ones is worrying the academic community. These challenges show how the lack of appr...The notoriety of the shortage of qualified professionals in the engineering segment to meet the existing projects and also the future ones is worrying the academic community. These challenges show how the lack of appropriate courses and low expenses with incentives to research and extension programs can affect the formation of the future engineer. Therefore, universities have the mission to develop teaching, research and extension, offering to the students new opportunities for diverse technical training, scientific and humanist formation. It is noted, however, that such activities in many engineering courses, especially scientific research, are not being prioritized by the universities. In light of this, the present paper aims to register measure and evaluate the participation of the students in scientific initiation in the four engineering courses of the Faculty of Engineering of the Minas Gerais State University. Sticking to the disparities presented by the four courses studied, in relation to the participation in research projects, the results showed a greater engagement of students of Environmental Engineering and Mining Engineering courses regarding the other engineering courses. In addition, a better divulgation and a greater involvement of teachers in projects were identified as the main recurring challenges to the access in scientific research by the students of this institution.展开更多
Cation segregation on cathode surfaces plays a key role in determining the activity and operational stability of solid oxide fuel cells(SOFCs).The double perovskite oxide PrBa_(0.8)Ca_(0.2)Co_(2)O_(5+δ)(PBCC)has been...Cation segregation on cathode surfaces plays a key role in determining the activity and operational stability of solid oxide fuel cells(SOFCs).The double perovskite oxide PrBa_(0.8)Ca_(0.2)Co_(2)O_(5+δ)(PBCC)has been widely studied as an active cathode but still suffer from serious detrimental segregations.To enhance the cathode stability,a PBCC derived A-site medium-entropy Pr_(0.6)La_(0.1)Nd_(0.1)Sm_(0.1)Gd_(0.1)Ba_(0.8)Ca_(0.2)Co_(2)O_(5+δ)(ME-PBCC)oxide was prepared and its segregation behaviors were investigated under different conditions.Compared with initial PBCC oxide,the segregations of BaO and Co_(3)O_(4)on the surface of ME-PBCC material are significantly suppressed,especially for Co_(3)O_(4),which is attributed to its higher configuration entropy.Our results also confirm the improved electrochemical performance and structural stability of ME-PBCC material,enabling it as a promising cathode for SOFCs.展开更多
Magnesium(Mg)-based bioresorbable stents represent a potentially groundbreaking advancement in cardiovascular therapy;offering tem-porary vessel support and complete biodegradability—addressing limitations of traditi...Magnesium(Mg)-based bioresorbable stents represent a potentially groundbreaking advancement in cardiovascular therapy;offering tem-porary vessel support and complete biodegradability—addressing limitations of traditional stents like in-stent restenosis and long-term com-plications.However,challenges such as rapid corrosion and suboptimal endothelialisation have hindered their clinical adoption.This review highlights the latest breakthroughs in surface modification,alloying,and coating strategies to enhance the mechanical integrity,corrosion resistance,and biocompatibility of Mg-based stents.Key surface engineering techniques,including polymer and bioactive coatings,are ex-amined for their role in promoting endothelial healing and minimising inflammatory responses.Future directions are proposed,focusing on personalised stent designs to optimize efficacy and long-term outcomes,positioning Mg-based stents as a transformative solution in interventional cardiology.展开更多
Dielectric materials are essential in modern electronics,serving as the backbone of numerous components across a wide array of electronic devices[1,2].As technology advances,the demand for materials with high permitti...Dielectric materials are essential in modern electronics,serving as the backbone of numerous components across a wide array of electronic devices[1,2].As technology advances,the demand for materials with high permittivity,low dielectric loss,and thermal stability continues to rise.Traditional strategies to enhance permittivity often involve mechanisms such as phase transitions in ferroelectrics or interfacial polarization in boundary layer capacitor(IBLC)systems.However,each comes with trade-offs.展开更多
This study provides the first systematic evaluation of image resolution’s effect (50-300 PPI, pixels per inch) on UAV (unmanned aerial vehicle)-based digital close-range photogrammetry accuracy in civil engineering a...This study provides the first systematic evaluation of image resolution’s effect (50-300 PPI, pixels per inch) on UAV (unmanned aerial vehicle)-based digital close-range photogrammetry accuracy in civil engineering applications, such as infrastructure monitoring and heritage preservation. Using a high-resolution UAV with a 20 MP (MegaPixels) sensor, four images of a brick wall test field were captured and processed in Agisoft Metashape, with resolutions compared against Leica T2002 theodolite measurements (1.0 mm accuracy). Advanced statistical methods (ANOVA (analysis of variance), Tukey tests, Monte Carlo simulations) and ground control points validated the results. Accuracy improved from 25 mm at 50 PPI to 5 mm at 150 PPI (p < 0.01), plateauing at 4 mm beyond 200 PPI, while 150 PPI reduced processing time by 62% compared to 300 PPI. Unlike prior studies, this research uniquely isolates resolution effects in a controlled civil engineering context, offering a novel 150 PPI threshold that balances precision and efficiency. This threshold supports Saudi Vision 2030’s smart infrastructure goals for megaprojects like NEOM, providing a scalable framework for global applications. Future research should leverage deep learning to optimize resolutions in dynamic environments.展开更多
Cardiac tissue engineering aims to efficiently replace or repair injured heart tissue using scaffolds,relevant cells,or their combination.While the combination of scaffolds and relevant cells holds the potential to ra...Cardiac tissue engineering aims to efficiently replace or repair injured heart tissue using scaffolds,relevant cells,or their combination.While the combination of scaffolds and relevant cells holds the potential to rapidly remuscularize the heart,thereby avoiding the slow process of cell recruitment,the proper ex vivo cellularization of a scaffold poses a substantial challenge.First,proper diffusion of nutrients and oxygen should be provided to the cell-seeded scaffold.Second,to generate a functional tissue construct,cells can benefit from physiological-like conditions.To meet these challenges,we developed a modular bioreactor for the dynamic cellularization of full-thickness cardiac scaffolds under synchronized mechanical and electrical stimuli.In this unique bioreactor system,we designed a cyclic mechanical load that mimics the left ventricle volume inflation,thus achieving a steady stimulus,as well as an electrical stimulus with an action potential profile to mirror the cells’microenvironment and electrical stimuli in the heart.These mechanical and electrical stimuli were synchronized according to cardiac physiology and regulated by constant feedback.When applied to a seeded thick porcine cardiac extracellular matrix(pcECM)scaffold,these stimuli improved the proliferation of mesenchymal stem/stromal cells(MSCs)and induced the formation of a dense tissue-like structure near the scaffold’s surface.Most importantly,after 35 d of cultivation,the MSCs presented the early cardiac progenitor markers Connexin-43 andα-actinin,which were absent in the control cells.Overall,this research developed a new bioreactor system for cellularizing cardiac scaffolds under cardiac-like conditions,aiming to restore a sustainable dynamic living tissue that can bear the essential cardiac excitation–contraction coupling.展开更多
Fatigue failure continues to be a significant challenge in designing structural and mechanical components subjected to repeated and complex loading.While earlier studies mainly examined material properties and how str...Fatigue failure continues to be a significant challenge in designing structural and mechanical components subjected to repeated and complex loading.While earlier studies mainly examined material properties and how stress affects lifespan,this review offers the first comprehensive,multiscale comparison of strategies that optimize geometry to improve fatigue performance.This includes everything from microscopic features like the shape of graphite nodules to large-scale design elements such as fillets,notches,and overall structural layouts.We analyze and combine various methods,including topology and shape optimization,the ability of additive manufacturing to finetune internal geometries,and reliability-based design approaches.A key new contribution is our proposal of a standard way to evaluate geometry-focused fatigue design,allowing for consistent comparison and encouraging validation across different fields.Furthermore,we highlight important areas for future research,such as incorporating manufacturing flaws,using multiscale models,and integrating machine learning techniques.This work is the first to provide a broad geometric viewpoint in fatigue engineering,laying the groundwork for future design methods that are driven by data and centered on reliability.展开更多
Transition metal oxides(TMOs),thanks to their elevated theoretical capacitance and pseudocapacitive properties,are of particular interest in exploring the advanced supercapacitor electrode materials.The present work r...Transition metal oxides(TMOs),thanks to their elevated theoretical capacitance and pseudocapacitive properties,are of particular interest in exploring the advanced supercapacitor electrode materials.The present work reports the rapid laser-assisted synthesis of SiC@-Fe_(2)O_(3-x)anode materials with engineered oxygen vacancies in seconds,which improve the charge transport,redox activity,and structural stability,thus facilitating a substantial enhancement in electrochemical performance.As a result,the resultant SiC@Fe_(2)O_(3-x)nanowires exhibit excellent performances with an areal capacitance of 1082.16 at 5 mA cm^(-2),and retain 86.7%capacitance over 10,000 cycles.Furthermore,the assembled asymmetric supercapacitors(ASC),employing SiC@Fe_(2)O_(3-x)as the negative electrode and Ni(OH)2as the positive electrode,delivers a 1.5 V operating voltage,an energy density of 197μWh cm^(-2),and 80.6%capacitance retention after 14,000cycles,representing their promise toward the applications in next-generation energy storage materials.展开更多
Background:Acquiring relevant information about procurement targets is fundamental to procuring medical devices.Although traditional Natural Language Processing(NLP)and Machine Learning(ML)methods have improved inform...Background:Acquiring relevant information about procurement targets is fundamental to procuring medical devices.Although traditional Natural Language Processing(NLP)and Machine Learning(ML)methods have improved information retrieval efficiency to a certain extent,they exhibit significant limitations in adaptability and accuracy when dealing with procurement documents characterized by diverse formats and a high degree of unstructured content.The emergence of Large Language Models(LLMs)offers new possibilities for efficient procurement information processing and extraction.Methods:This study collected procurement transaction documents from public procurement websites,and proposed a procurement Information Extraction(IE)method based on LLMs.Unlike traditional approaches,this study systematically explores the applicability of LLMs in both structured and unstructured entities in procurement documents,addressing the challenges posed by format variability and content complexity.Furthermore,an optimized prompt framework tailored for procurement document extraction tasks is developed to enhance the accuracy and robustness of IE.The aim is to process and extract key information from medical device procurement quickly and accurately,meeting stakeholders'demands for precision and timeliness in information retrieval.Results:Experimental results demonstrate that,compared to traditional methods,the proposed approach achieves an F1 Score of 0.9698,representing a 4.85%improvement over the best baseline model.Moreover,both recall and precision rates are close to 97%,significantly outperforming other models and exhibiting exceptional overall recognition capabilities.Notably,further analysis reveals that the proposed method consistently maintains high performance across both structured and unstructured entities in procurement documents while balancing recall and precision effectively,demonstrating its adaptability in handling varying document formats.The results of ablation experiments validate the effectiveness of the proposed prompting strategy.Conclusion:Additionally,this study explores the challenges and potential improvements of the proposed method in IE tasks and provides insights into its feasibility for real-world deployment and application directions,further clarifying its adaptability and value.This method not only exhibits significant advantages in medical device procurement but also holds promise for providing new approaches to information processing and decision support in various domains.展开更多
This study introduces the type-I heavy-tailed Burr XII(TIHTBXII)distribution,a highly flexible and robust statistical model designed to address the limitations of conventional distributions in analyzing data character...This study introduces the type-I heavy-tailed Burr XII(TIHTBXII)distribution,a highly flexible and robust statistical model designed to address the limitations of conventional distributions in analyzing data characterized by skewness,heavy tails,and diverse hazard behaviors.We meticulously develop the TIHTBXII’s mathematical foundations,including its probability density function(PDF),cumulative distribution function(CDF),and essential statistical properties,crucial for theoretical understanding and practical application.A comprehensive Monte Carlo simulation evaluates four parameter estimation methods:maximum likelihood(MLE),maximum product spacing(MPS),least squares(LS),and weighted least squares(WLS).The simulation results consistently show that as sample sizes increase,the Bias and RMSE of all estimators decrease,with WLS and LS often demonstrating superior and more stable performance.Beyond theoretical development,we present a practical application of the TIHTBXII distribution in constructing a group acceptance sampling plan(GASP)for truncated life tests.This application highlights how the TIHTBXII model can optimize quality control decisions by minimizing the average sample number(ASN)while effectively managing consumer and producer risks.Empirical validation using real-world datasets,including“Active Repair Duration,”“Groundwater Contaminant Measurements,”and“Dominica COVID-19 Mortality,”further demonstrates the TIHTBXII’s superior fit compared to existing models.Our findings confirm the TIHTBXII distribution as a powerful and reliable alternative for accurately modeling complex data in fields such as reliability engineering and quality assessment,leading to more informed and robust decision-making.展开更多
The limited ion/electron transport kinetics and insufficient crystalline stability of TiNb_(2)O_(7)(TNO)present significant challenges to the development of high-performance lithium-ion batteries(LIBs)with fastchargin...The limited ion/electron transport kinetics and insufficient crystalline stability of TiNb_(2)O_(7)(TNO)present significant challenges to the development of high-performance lithium-ion batteries(LIBs)with fastcharging capabilities and long cycle life.Here we propose a dual-modification strategy combining Ndoped carbon(NC)coating and Co^(2+)/W^(6+)doping,which not only enhances ionic and electronic conductivity but also effectively regulates volume expansion during electrochemical cycling.Upon Li+ion insertion,a significant reduction in the unit cell expansion coefficient of doped TNO is observed,from 7.48%(pristine TNO)to 5.37%(with 3%W^(6+)doping)and 4.65%(with 3%Co^(2+)doping),alo ng with lowered lattice distortion and improved uniformity in internal strain release.Density functional theory(DFT)simulation demonstrates that Co^(2+)and W^(6+)ions preferentially substitute Ti^(4+)sites in the TNO crystal,leading to the improved electronic conductivity by narrowing the bandgap.Moreover,Co^(2+)doping creates lower electron density and wider Li+ion transport channels than W^(6+)doping.The optimized 3Co-TNO@NC anode delivers a remarkable power density of 11.0 kW kg^(-1)at 20 C while maintaining a high reversible capacity of 150.9 mAh g^(-1)at 10 C after 2000 cycles.It also exhibits excellent compatibility in full cells,paired well with LiFePO_(4)(137.9 mAh g^(-1)after 2000 cycles)and Ni-rich LiNi_(0.8)Co_(0.1)Mn_(0.1)(130.9 mAh g^(-1)after 500cycles)cathodes at 5 C,highlighting its potential as a high-safety,low-strain anode material for highpower LIBs.展开更多
Lithium–sulfur(Li–S)batteries are promisingcandidates for next-generation energy storagegiven their high energy density and potential low cost.Chemically activated carbon(CAC)is often used fortheir cathodes,because ...Lithium–sulfur(Li–S)batteries are promisingcandidates for next-generation energy storagegiven their high energy density and potential low cost.Chemically activated carbon(CAC)is often used fortheir cathodes,because it has a high specific surfacearea for sulfur loading.We have developed a pressurizedphysical activation(PPA)method that producedan activated carbon(PPAC)with a high specific surfacearea comparable to that of CAC.The pore structure of PPAC could be changed and its use as a cathode material for Li–Sbatteries was investigated.Battery tests at different capacity rates(C-rates)showed that it had a much improved high-rate performancewith a discharge capacity of 900 mAh/(g of sulfur)at 1 C,in contrast to only 600 mAh/(g of sulfur)for CAC.Porestructure analyses showed that PPAC prepared at a high activation temperature(1000℃)had unusual channel-like mesoporesbetween the microdomains that are the basic structural units of artificial carbon materials.These are connected to microporesdeveloped in each microdomain,and deliver ions from the surroundings to the internal pores and vice versa.The well-developedmicropores and mesopores of PPAC respectively ensured the high adsorption of lithium polysulfides and a high rate ofion diffusion.Compared to CAC,PPAC is a high-performance,low-cost cathode material that is promising for use in futureLi–S batteries.展开更多
Soft robotic manipulators represent a rapidly evolving field characterized by inherent compliance,adaptability,and safe interactions within unstructured environments.Over the past decade(2015-2025),significant advance...Soft robotic manipulators represent a rapidly evolving field characterized by inherent compliance,adaptability,and safe interactions within unstructured environments.Over the past decade(2015-2025),significant advancements have trans-formed their capabilities through novel designs inspired by biological systems,advanced modeling frameworks,sophisti-cated control strategies,and integration into diverse real-world applications.Recent innovations in multifunctional mate-rials and emerging actuation technologies have markedly expanded manipulator performance,reliability,and dexterity.Concurrently,developments in modeling have progressed from simplified geometric methods toward highly accurate physics-based and hybrid data-driven approaches,substantially improving real-time prediction and controllability.Coupled with these developments,adaptive and robust control strategies-including learning-based techniques-have enabled unprec-edented autonomy and precision in challenging application domains such as Minimally Invasive Surgery(MIS),precision agriculture,deep-sea exploration,disaster recovery,and space missions.Despite these remarkable strides,key challenges remain,notably regarding scalability,long-term material durability,robust integrated sensing,and standardized evaluation procedures.This review comprehensively synthesizes recent advances,critically evaluates state-of-the-art methodologies,and systematically identifies existing gaps to provide a clear roadmap and targeted research directions,guiding future developments toward the broader adoption and optimal utilization of soft robotic manipulators.展开更多
Generalised reduced masses with a set of equations governing the three relative motions between two of 3-bodies in their gravitational field are established,of which the dynamic characteristics of 3-body dynamics,fund...Generalised reduced masses with a set of equations governing the three relative motions between two of 3-bodies in their gravitational field are established,of which the dynamic characteristics of 3-body dynamics,fundamental bases of this paper,are revealed.Based on these findings,an equivalent system is developed,which is a 2-body system with its total mass,constant angular momentum,kinetic and potential energies same as the total ones of three relative motions,so that it can be solved using the well-known theory of the 2-body system.From the solution of an equivalent system with the revealed characteristics of three relative motions,the general theoretical solutions of the 3-body system are obtained in the curve-integration forms along the orbits in the imaged radial motion space.The possible periodical orbits with generalised Kepler’s law are presented.Following the description and mathematical demonstrations of the proposed methods,the examples including Euler’s/Lagrange’s problems,and a reported numerical one are solved to validate the proposed methods.The methods derived from the 3-body system are extended to N-body problems.展开更多
Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Re...Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Recent approaches such as Distilling Step-by-Step(DSbS)introduce explanation supervision,yet they apply it in a uniform manner that may not fully exploit the different learning dynamics of prediction and explanation.In this work,we propose a task-structured curriculum learning(TSCL)framework that structures training into three sequential phases:(i)prediction-only,to establish stable feature representations;(ii)joint prediction-explanation,to align task outputs with rationale generation;and(iii)explanation-only,to refine the quality of rationales.This design provides a simple but effective modification to DSbS,requiring no architectural changes and adding negligible training cost.We justify the phase scheduling with ablation studies and convergence analysis,showing that an initial prediction-heavy stage followed by a balanced joint phase improves both stability and explanation alignment.Extensive experiments on five datasets(e-SNLI,ANLI,CommonsenseQA,SVAMP,and MedNLI)demonstrate that TSCL consistently outperforms strong baselines,achieving gains of+1.7-2.6 points in accuracy and 0.8-1.2 in ROUGE-L,corresponding to relative error reductions of up to 21%.Beyond lexical metrics,human evaluation and ERASERstyle faithfulness diagnostics confirm that TSCL produces more faithful and informative explanations.Comparative training curves further reveal faster convergence and lower variance across seeds.Efficiency analysis shows less than 3%overhead in wall-clock training time and no additional inference cost,making the approach practical for realworld deployment.This study demonstrates that a simple task-structured curriculum can significantly improve the effectiveness of knowledge distillation.By separating and sequencing objectives,TSCL achieves a better balance between accuracy,stability,and explanation quality.The framework generalizes across domains,including medical NLI,and offers a principled recipe for future applications in multimodal reasoning and reinforcement learning.展开更多
Neuronanomedicine is a promising interdisciplinary field combining two critical fields,neuroscience and nanotechnology.This study focuses on the engineering of magnetized nanoparticles(MNPs)in diagnosing and treating ...Neuronanomedicine is a promising interdisciplinary field combining two critical fields,neuroscience and nanotechnology.This study focuses on the engineering of magnetized nanoparticles(MNPs)in diagnosing and treating neurological disorders and brain cancer.Additionally,this mechanism enhances the effectiveness of magnetic-guided drug delivery.The alternating magnetic field is applied to control the directions of the MNPs to target the tumor cells.This study approaches the radiotherapy techniques of magnetic hyperthermia therapy(MHT),wherein the thermal radiative heat transfer effect is applied to achieve homogenous heating to destroy cancer cells.MNPs are injected through the cerebrospinal fluid(CSF)transport in the glymphatic system.The elastic properties of the cerebral arteries cause peristaltic propulsion for the resulting nanofluid.Therefore,the effective Maxwell model for the nanofluid thermal conductivity is selected.The nanofluid governing equations are solved using the perturbation technique under small wavelength number and long wavelength approximation with small Reynolds number.Additionally,the effects of thermal slip and elastic properties boundary conditions are incorporated.The graphical results for the streamwise velocity,pressure,and temperature distributions are plotted using MATLAB package considering the different effects of the magnetic flux intensity,thermal radiation parameter,thermal slipping at boundaries,elastic wall properties,and nanoparticle concentration.The results demonstrate the strong impact of the magnetic field and radiation heating in terms of enhancing the nanofluid CSF flow behavior and destroying cancer.展开更多
In this work,a computational modelling and analysis framework is developed to investigate the thermal buckling behavior of doubly-curved composite shells reinforced with graphene-origami(G-Ori)auxetic metamaterials.A ...In this work,a computational modelling and analysis framework is developed to investigate the thermal buckling behavior of doubly-curved composite shells reinforced with graphene-origami(G-Ori)auxetic metamaterials.A semi-analytical formulation based on the First-Order Shear Deformation Theory(FSDT)and the principle of virtual displacements is established,and closed-form solutions are derived via Navier’s method for simply supported boundary conditions.The G-Ori metamaterial reinforcements are treated as programmable constructs whose effective thermo-mechanical properties are obtained via micromechanical homogenization and incorporated into the shell model.A comprehensive parametric study examines the influence of folding geometry,dispersion arrangement,reinforcement weight fraction,curvature parameters,and elastic foundation support on the critical buckling temperature(CBT).The results reveal that,under optimal folding geometry and reinforcement alignment with principal stress trajectories,the CBT can increase by more than 150%.Furthermore,the combined effect of G-Ori reinforcement and elastic foundation substantially enhances thermal buckling resistance.These findings establish design guidelines for architected composite shells in applications such as aerospace thermal skins,morphing structures,and thermally-responsive systems,and illustrate the potential of auxetic graphene metamaterials for multifunctional,lightweight,and thermally robust structural components.展开更多
Breast cancer screening programs rely heavily on mammography for early detection;however,diagnostic performance is strongly affected by inter-reader variability,breast density,and the limitations of conven-tional comp...Breast cancer screening programs rely heavily on mammography for early detection;however,diagnostic performance is strongly affected by inter-reader variability,breast density,and the limitations of conven-tional computer-aided detection systems.Recent advances in deep learning have enabled more robust and scalable solutions for large-scale screening,yet a systematic comparison of modern object detection architectures on nationally representative datasets remains limited.This study presents a comprehensive quantitative comparison of prominent deep learning–based object detection architectures for Artificial Intelligence-assisted mammography analysis using the MammosighTR dataset,developed within the Turkish National Breast Cancer Screening Program.The dataset comprises 12,740 patient cases collected between 2016 and 2022,annotated with BI-RADS categories,breast density levels,and lesion localization labels.A total of 31 models were evaluated,including One-Stage,Two-Stage,and Transformer-based architectures,under a unified experimental framework at both patient and breast levels.The results demonstrate that Two-Stage architectures consistently outperform One-Stage models,achieving approximately 2%–4%higher Macro F1-Scores and more balanced precision–recall trade-offs,with Double-Head R-CNN and Dynamic R-CNN yielding the highest overall performance(Macro F1≈0.84–0.86).This advantage is primarily attributed to the region proposal mechanism and improved class balance inherent to Two-Stage designs.One-Stage detectors exhibited higher sensitivity and faster inference,reaching Recall values above 0.88,but experienced minor reductions in Precision and overall accuracy(≈1%–2%)compared with Two-Stage models.Among Transformer-based architectures,Deformable DEtection TRansformer demonstrated strong robustness and consistency across datasets,achieving Macro F1-Scores comparable to CNN-based detectors(≈0.83–0.85)while exhibiting minimal performance degradation under distributional shifts.Breast density–based analysis revealed increased misclassification rates in medium-density categories(types B and C),whereas Transformer-based architectures maintained more stable performance in high-density type D tissue.These findings quantitatively confirm that both architectural design and tissue characteristics play a decisive role in diagnostic accuracy.Overall,the study provides a reproducible benchmark and highlights the potential of hybrid approaches that combine the accuracy of Two-Stage detectors with the contextual modeling capability of Transformer architectures for clinically reliable breast cancer screening systems.展开更多
Accurate short-term electricity price forecasts are essential for market participants to optimize bidding strategies,hedge risk and plan generation schedules.By leveraging advanced data analytics and machine learning ...Accurate short-term electricity price forecasts are essential for market participants to optimize bidding strategies,hedge risk and plan generation schedules.By leveraging advanced data analytics and machine learning methods,accurate and reliable price forecasts can be achieved.This study forecasts day-ahead prices in Türkiye’s electricity market using eXtreme Gradient Boosting(XGBoost).We benchmark XGBoost against four alternatives—Support Vector Machines(SVM),Long Short-Term Memory(LSTM),Random Forest(RF),and Gradient Boosting(GBM)—using 8760 hourly observations from 2023 provided by Energy Exchange Istanbul(EXIST).All models were trained on an identical chronological 80/20 train–test split,with hyperparameters tuned via 5-fold cross-validation on the training set.XGBoost achieved the best performance(Mean Absolute Error(MAE)=144.8 TRY/MWh,Root Mean Square Error(RMSE)=201.8 TRY/MWh,coefficient of determination(R^(2))=0.923)while training in 94 s.To enhance interpretability and identify key drivers,we employed Shapley Additive Explanations(SHAP),which highlighted a strong association between higher prices and increased natural-gas-based generation.The results provide a clear performance benchmark and practical guidance for selecting forecasting approaches in day-ahead electricity markets.展开更多
文摘The study was conducted to identify indoor air quality and the level of thermal comfort in various selected locations in Faculty of Engineering and Built Environment （FKAB）, University Kebangsaan Malaysia （UKM） with built-up area of 250,936 fie. The indoor air quality and thermal comfort were measured at various selected locations by using indoor air quality equipment （Thermal Comfort SERI）. The thermal comfort assessments are based on Malaysian Code of Practice Indoor Air Quality 2005 and Moderate Thermal Environments-Determination of the PMV and PPD indices specification of the condition for thermal comfort （ISO7730：1994） From the data analysis, the FKAB building is considered inadequately vented space. The concentration of CO2 for all sampling area evaluated exceeds the recommended concentration （〉 1000 ppm）. The ventilation system used in FKAB building is designed by delivering fix amount of fresh air into building from external building without consideration on the number of occupants. This common ventilation design will increase the amount of CO2 dramatically all day long and these reflect the inefficiency of energy used. The faculty needs to be equipped with a comprehensive energy management system that can allow detailed documentation of continuous performance of all energy system and consumption in the building.
文摘The notoriety of the shortage of qualified professionals in the engineering segment to meet the existing projects and also the future ones is worrying the academic community. These challenges show how the lack of appropriate courses and low expenses with incentives to research and extension programs can affect the formation of the future engineer. Therefore, universities have the mission to develop teaching, research and extension, offering to the students new opportunities for diverse technical training, scientific and humanist formation. It is noted, however, that such activities in many engineering courses, especially scientific research, are not being prioritized by the universities. In light of this, the present paper aims to register measure and evaluate the participation of the students in scientific initiation in the four engineering courses of the Faculty of Engineering of the Minas Gerais State University. Sticking to the disparities presented by the four courses studied, in relation to the participation in research projects, the results showed a greater engagement of students of Environmental Engineering and Mining Engineering courses regarding the other engineering courses. In addition, a better divulgation and a greater involvement of teachers in projects were identified as the main recurring challenges to the access in scientific research by the students of this institution.
基金Project supported by the National Natural Science Foundation of China(22279025,21773048,52302119)the Fundamental Research Funds for the Central Universities(2023FRFK06005,HIT.NSRIF202204)。
文摘Cation segregation on cathode surfaces plays a key role in determining the activity and operational stability of solid oxide fuel cells(SOFCs).The double perovskite oxide PrBa_(0.8)Ca_(0.2)Co_(2)O_(5+δ)(PBCC)has been widely studied as an active cathode but still suffer from serious detrimental segregations.To enhance the cathode stability,a PBCC derived A-site medium-entropy Pr_(0.6)La_(0.1)Nd_(0.1)Sm_(0.1)Gd_(0.1)Ba_(0.8)Ca_(0.2)Co_(2)O_(5+δ)(ME-PBCC)oxide was prepared and its segregation behaviors were investigated under different conditions.Compared with initial PBCC oxide,the segregations of BaO and Co_(3)O_(4)on the surface of ME-PBCC material are significantly suppressed,especially for Co_(3)O_(4),which is attributed to its higher configuration entropy.Our results also confirm the improved electrochemical performance and structural stability of ME-PBCC material,enabling it as a promising cathode for SOFCs.
文摘Magnesium(Mg)-based bioresorbable stents represent a potentially groundbreaking advancement in cardiovascular therapy;offering tem-porary vessel support and complete biodegradability—addressing limitations of traditional stents like in-stent restenosis and long-term com-plications.However,challenges such as rapid corrosion and suboptimal endothelialisation have hindered their clinical adoption.This review highlights the latest breakthroughs in surface modification,alloying,and coating strategies to enhance the mechanical integrity,corrosion resistance,and biocompatibility of Mg-based stents.Key surface engineering techniques,including polymer and bioactive coatings,are ex-amined for their role in promoting endothelial healing and minimising inflammatory responses.Future directions are proposed,focusing on personalised stent designs to optimize efficacy and long-term outcomes,positioning Mg-based stents as a transformative solution in interventional cardiology.
文摘Dielectric materials are essential in modern electronics,serving as the backbone of numerous components across a wide array of electronic devices[1,2].As technology advances,the demand for materials with high permittivity,low dielectric loss,and thermal stability continues to rise.Traditional strategies to enhance permittivity often involve mechanisms such as phase transitions in ferroelectrics or interfacial polarization in boundary layer capacitor(IBLC)systems.However,each comes with trade-offs.
文摘This study provides the first systematic evaluation of image resolution’s effect (50-300 PPI, pixels per inch) on UAV (unmanned aerial vehicle)-based digital close-range photogrammetry accuracy in civil engineering applications, such as infrastructure monitoring and heritage preservation. Using a high-resolution UAV with a 20 MP (MegaPixels) sensor, four images of a brick wall test field were captured and processed in Agisoft Metashape, with resolutions compared against Leica T2002 theodolite measurements (1.0 mm accuracy). Advanced statistical methods (ANOVA (analysis of variance), Tukey tests, Monte Carlo simulations) and ground control points validated the results. Accuracy improved from 25 mm at 50 PPI to 5 mm at 150 PPI (p < 0.01), plateauing at 4 mm beyond 200 PPI, while 150 PPI reduced processing time by 62% compared to 300 PPI. Unlike prior studies, this research uniquely isolates resolution effects in a controlled civil engineering context, offering a novel 150 PPI threshold that balances precision and efficiency. This threshold supports Saudi Vision 2030’s smart infrastructure goals for megaprojects like NEOM, providing a scalable framework for global applications. Future research should leverage deep learning to optimize resolutions in dynamic environments.
基金funded by the Israeli Ministry of Innovation,Science and Technology(Grant No.3-11873)the Israel Science Foundation(Grant No.1563/10)+1 种基金the Randy L.and Melvin R.Berlin Family Research Center for Regenerative Medicinethe Gurwin Family Foundation.
文摘Cardiac tissue engineering aims to efficiently replace or repair injured heart tissue using scaffolds,relevant cells,or their combination.While the combination of scaffolds and relevant cells holds the potential to rapidly remuscularize the heart,thereby avoiding the slow process of cell recruitment,the proper ex vivo cellularization of a scaffold poses a substantial challenge.First,proper diffusion of nutrients and oxygen should be provided to the cell-seeded scaffold.Second,to generate a functional tissue construct,cells can benefit from physiological-like conditions.To meet these challenges,we developed a modular bioreactor for the dynamic cellularization of full-thickness cardiac scaffolds under synchronized mechanical and electrical stimuli.In this unique bioreactor system,we designed a cyclic mechanical load that mimics the left ventricle volume inflation,thus achieving a steady stimulus,as well as an electrical stimulus with an action potential profile to mirror the cells’microenvironment and electrical stimuli in the heart.These mechanical and electrical stimuli were synchronized according to cardiac physiology and regulated by constant feedback.When applied to a seeded thick porcine cardiac extracellular matrix(pcECM)scaffold,these stimuli improved the proliferation of mesenchymal stem/stromal cells(MSCs)and induced the formation of a dense tissue-like structure near the scaffold’s surface.Most importantly,after 35 d of cultivation,the MSCs presented the early cardiac progenitor markers Connexin-43 andα-actinin,which were absent in the control cells.Overall,this research developed a new bioreactor system for cellularizing cardiac scaffolds under cardiac-like conditions,aiming to restore a sustainable dynamic living tissue that can bear the essential cardiac excitation–contraction coupling.
文摘Fatigue failure continues to be a significant challenge in designing structural and mechanical components subjected to repeated and complex loading.While earlier studies mainly examined material properties and how stress affects lifespan,this review offers the first comprehensive,multiscale comparison of strategies that optimize geometry to improve fatigue performance.This includes everything from microscopic features like the shape of graphite nodules to large-scale design elements such as fillets,notches,and overall structural layouts.We analyze and combine various methods,including topology and shape optimization,the ability of additive manufacturing to finetune internal geometries,and reliability-based design approaches.A key new contribution is our proposal of a standard way to evaluate geometry-focused fatigue design,allowing for consistent comparison and encouraging validation across different fields.Furthermore,we highlight important areas for future research,such as incorporating manufacturing flaws,using multiscale models,and integrating machine learning techniques.This work is the first to provide a broad geometric viewpoint in fatigue engineering,laying the groundwork for future design methods that are driven by data and centered on reliability.
基金financially supported by the National Natural Science Foundation of China(No.52372063)China Postdoctoral Science Foundation(No.2023M730391)
文摘Transition metal oxides(TMOs),thanks to their elevated theoretical capacitance and pseudocapacitive properties,are of particular interest in exploring the advanced supercapacitor electrode materials.The present work reports the rapid laser-assisted synthesis of SiC@-Fe_(2)O_(3-x)anode materials with engineered oxygen vacancies in seconds,which improve the charge transport,redox activity,and structural stability,thus facilitating a substantial enhancement in electrochemical performance.As a result,the resultant SiC@Fe_(2)O_(3-x)nanowires exhibit excellent performances with an areal capacitance of 1082.16 at 5 mA cm^(-2),and retain 86.7%capacitance over 10,000 cycles.Furthermore,the assembled asymmetric supercapacitors(ASC),employing SiC@Fe_(2)O_(3-x)as the negative electrode and Ni(OH)2as the positive electrode,delivers a 1.5 V operating voltage,an energy density of 197μWh cm^(-2),and 80.6%capacitance retention after 14,000cycles,representing their promise toward the applications in next-generation energy storage materials.
文摘Background:Acquiring relevant information about procurement targets is fundamental to procuring medical devices.Although traditional Natural Language Processing(NLP)and Machine Learning(ML)methods have improved information retrieval efficiency to a certain extent,they exhibit significant limitations in adaptability and accuracy when dealing with procurement documents characterized by diverse formats and a high degree of unstructured content.The emergence of Large Language Models(LLMs)offers new possibilities for efficient procurement information processing and extraction.Methods:This study collected procurement transaction documents from public procurement websites,and proposed a procurement Information Extraction(IE)method based on LLMs.Unlike traditional approaches,this study systematically explores the applicability of LLMs in both structured and unstructured entities in procurement documents,addressing the challenges posed by format variability and content complexity.Furthermore,an optimized prompt framework tailored for procurement document extraction tasks is developed to enhance the accuracy and robustness of IE.The aim is to process and extract key information from medical device procurement quickly and accurately,meeting stakeholders'demands for precision and timeliness in information retrieval.Results:Experimental results demonstrate that,compared to traditional methods,the proposed approach achieves an F1 Score of 0.9698,representing a 4.85%improvement over the best baseline model.Moreover,both recall and precision rates are close to 97%,significantly outperforming other models and exhibiting exceptional overall recognition capabilities.Notably,further analysis reveals that the proposed method consistently maintains high performance across both structured and unstructured entities in procurement documents while balancing recall and precision effectively,demonstrating its adaptability in handling varying document formats.The results of ablation experiments validate the effectiveness of the proposed prompting strategy.Conclusion:Additionally,this study explores the challenges and potential improvements of the proposed method in IE tasks and provides insights into its feasibility for real-world deployment and application directions,further clarifying its adaptability and value.This method not only exhibits significant advantages in medical device procurement but also holds promise for providing new approaches to information processing and decision support in various domains.
基金supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University(IMSIU)(Grant Number IMSIU-DDRSP2501).
文摘This study introduces the type-I heavy-tailed Burr XII(TIHTBXII)distribution,a highly flexible and robust statistical model designed to address the limitations of conventional distributions in analyzing data characterized by skewness,heavy tails,and diverse hazard behaviors.We meticulously develop the TIHTBXII’s mathematical foundations,including its probability density function(PDF),cumulative distribution function(CDF),and essential statistical properties,crucial for theoretical understanding and practical application.A comprehensive Monte Carlo simulation evaluates four parameter estimation methods:maximum likelihood(MLE),maximum product spacing(MPS),least squares(LS),and weighted least squares(WLS).The simulation results consistently show that as sample sizes increase,the Bias and RMSE of all estimators decrease,with WLS and LS often demonstrating superior and more stable performance.Beyond theoretical development,we present a practical application of the TIHTBXII distribution in constructing a group acceptance sampling plan(GASP)for truncated life tests.This application highlights how the TIHTBXII model can optimize quality control decisions by minimizing the average sample number(ASN)while effectively managing consumer and producer risks.Empirical validation using real-world datasets,including“Active Repair Duration,”“Groundwater Contaminant Measurements,”and“Dominica COVID-19 Mortality,”further demonstrates the TIHTBXII’s superior fit compared to existing models.Our findings confirm the TIHTBXII distribution as a powerful and reliable alternative for accurately modeling complex data in fields such as reliability engineering and quality assessment,leading to more informed and robust decision-making.
基金support from the BRICS STI Framework Programme(No.52261145703)National Research Foundation+2 种基金Singapore under Award No.NRF-CRP24-2020-0002the Italy-Singapore Science and Technology Cooperation(Grant No.R23101R040)the use of computing resources at the A*STAR Computational Centre and National Supercomputer Centre,Singapore。
文摘The limited ion/electron transport kinetics and insufficient crystalline stability of TiNb_(2)O_(7)(TNO)present significant challenges to the development of high-performance lithium-ion batteries(LIBs)with fastcharging capabilities and long cycle life.Here we propose a dual-modification strategy combining Ndoped carbon(NC)coating and Co^(2+)/W^(6+)doping,which not only enhances ionic and electronic conductivity but also effectively regulates volume expansion during electrochemical cycling.Upon Li+ion insertion,a significant reduction in the unit cell expansion coefficient of doped TNO is observed,from 7.48%(pristine TNO)to 5.37%(with 3%W^(6+)doping)and 4.65%(with 3%Co^(2+)doping),alo ng with lowered lattice distortion and improved uniformity in internal strain release.Density functional theory(DFT)simulation demonstrates that Co^(2+)and W^(6+)ions preferentially substitute Ti^(4+)sites in the TNO crystal,leading to the improved electronic conductivity by narrowing the bandgap.Moreover,Co^(2+)doping creates lower electron density and wider Li+ion transport channels than W^(6+)doping.The optimized 3Co-TNO@NC anode delivers a remarkable power density of 11.0 kW kg^(-1)at 20 C while maintaining a high reversible capacity of 150.9 mAh g^(-1)at 10 C after 2000 cycles.It also exhibits excellent compatibility in full cells,paired well with LiFePO_(4)(137.9 mAh g^(-1)after 2000 cycles)and Ni-rich LiNi_(0.8)Co_(0.1)Mn_(0.1)(130.9 mAh g^(-1)after 500cycles)cathodes at 5 C,highlighting its potential as a high-safety,low-strain anode material for highpower LIBs.
文摘Lithium–sulfur(Li–S)batteries are promisingcandidates for next-generation energy storagegiven their high energy density and potential low cost.Chemically activated carbon(CAC)is often used fortheir cathodes,because it has a high specific surfacearea for sulfur loading.We have developed a pressurizedphysical activation(PPA)method that producedan activated carbon(PPAC)with a high specific surfacearea comparable to that of CAC.The pore structure of PPAC could be changed and its use as a cathode material for Li–Sbatteries was investigated.Battery tests at different capacity rates(C-rates)showed that it had a much improved high-rate performancewith a discharge capacity of 900 mAh/(g of sulfur)at 1 C,in contrast to only 600 mAh/(g of sulfur)for CAC.Porestructure analyses showed that PPAC prepared at a high activation temperature(1000℃)had unusual channel-like mesoporesbetween the microdomains that are the basic structural units of artificial carbon materials.These are connected to microporesdeveloped in each microdomain,and deliver ions from the surroundings to the internal pores and vice versa.The well-developedmicropores and mesopores of PPAC respectively ensured the high adsorption of lithium polysulfides and a high rate ofion diffusion.Compared to CAC,PPAC is a high-performance,low-cost cathode material that is promising for use in futureLi–S batteries.
基金Open access funding provided by The Science,Technology&Innovation Funding Authority(STDF)in cooperation with The Egyptian Knowledge Bank(EKB).
文摘Soft robotic manipulators represent a rapidly evolving field characterized by inherent compliance,adaptability,and safe interactions within unstructured environments.Over the past decade(2015-2025),significant advancements have trans-formed their capabilities through novel designs inspired by biological systems,advanced modeling frameworks,sophisti-cated control strategies,and integration into diverse real-world applications.Recent innovations in multifunctional mate-rials and emerging actuation technologies have markedly expanded manipulator performance,reliability,and dexterity.Concurrently,developments in modeling have progressed from simplified geometric methods toward highly accurate physics-based and hybrid data-driven approaches,substantially improving real-time prediction and controllability.Coupled with these developments,adaptive and robust control strategies-including learning-based techniques-have enabled unprec-edented autonomy and precision in challenging application domains such as Minimally Invasive Surgery(MIS),precision agriculture,deep-sea exploration,disaster recovery,and space missions.Despite these remarkable strides,key challenges remain,notably regarding scalability,long-term material durability,robust integrated sensing,and standardized evaluation procedures.This review comprehensively synthesizes recent advances,critically evaluates state-of-the-art methodologies,and systematically identifies existing gaps to provide a clear roadmap and targeted research directions,guiding future developments toward the broader adoption and optimal utilization of soft robotic manipulators.
文摘Generalised reduced masses with a set of equations governing the three relative motions between two of 3-bodies in their gravitational field are established,of which the dynamic characteristics of 3-body dynamics,fundamental bases of this paper,are revealed.Based on these findings,an equivalent system is developed,which is a 2-body system with its total mass,constant angular momentum,kinetic and potential energies same as the total ones of three relative motions,so that it can be solved using the well-known theory of the 2-body system.From the solution of an equivalent system with the revealed characteristics of three relative motions,the general theoretical solutions of the 3-body system are obtained in the curve-integration forms along the orbits in the imaged radial motion space.The possible periodical orbits with generalised Kepler’s law are presented.Following the description and mathematical demonstrations of the proposed methods,the examples including Euler’s/Lagrange’s problems,and a reported numerical one are solved to validate the proposed methods.The methods derived from the 3-body system are extended to N-body problems.
文摘Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Recent approaches such as Distilling Step-by-Step(DSbS)introduce explanation supervision,yet they apply it in a uniform manner that may not fully exploit the different learning dynamics of prediction and explanation.In this work,we propose a task-structured curriculum learning(TSCL)framework that structures training into three sequential phases:(i)prediction-only,to establish stable feature representations;(ii)joint prediction-explanation,to align task outputs with rationale generation;and(iii)explanation-only,to refine the quality of rationales.This design provides a simple but effective modification to DSbS,requiring no architectural changes and adding negligible training cost.We justify the phase scheduling with ablation studies and convergence analysis,showing that an initial prediction-heavy stage followed by a balanced joint phase improves both stability and explanation alignment.Extensive experiments on five datasets(e-SNLI,ANLI,CommonsenseQA,SVAMP,and MedNLI)demonstrate that TSCL consistently outperforms strong baselines,achieving gains of+1.7-2.6 points in accuracy and 0.8-1.2 in ROUGE-L,corresponding to relative error reductions of up to 21%.Beyond lexical metrics,human evaluation and ERASERstyle faithfulness diagnostics confirm that TSCL produces more faithful and informative explanations.Comparative training curves further reveal faster convergence and lower variance across seeds.Efficiency analysis shows less than 3%overhead in wall-clock training time and no additional inference cost,making the approach practical for realworld deployment.This study demonstrates that a simple task-structured curriculum can significantly improve the effectiveness of knowledge distillation.By separating and sequencing objectives,TSCL achieves a better balance between accuracy,stability,and explanation quality.The framework generalizes across domains,including medical NLI,and offers a principled recipe for future applications in multimodal reasoning and reinforcement learning.
基金Fundación Mujeres por Africa for supporting this work through the fellowship awarded to her。
文摘Neuronanomedicine is a promising interdisciplinary field combining two critical fields,neuroscience and nanotechnology.This study focuses on the engineering of magnetized nanoparticles(MNPs)in diagnosing and treating neurological disorders and brain cancer.Additionally,this mechanism enhances the effectiveness of magnetic-guided drug delivery.The alternating magnetic field is applied to control the directions of the MNPs to target the tumor cells.This study approaches the radiotherapy techniques of magnetic hyperthermia therapy(MHT),wherein the thermal radiative heat transfer effect is applied to achieve homogenous heating to destroy cancer cells.MNPs are injected through the cerebrospinal fluid(CSF)transport in the glymphatic system.The elastic properties of the cerebral arteries cause peristaltic propulsion for the resulting nanofluid.Therefore,the effective Maxwell model for the nanofluid thermal conductivity is selected.The nanofluid governing equations are solved using the perturbation technique under small wavelength number and long wavelength approximation with small Reynolds number.Additionally,the effects of thermal slip and elastic properties boundary conditions are incorporated.The graphical results for the streamwise velocity,pressure,and temperature distributions are plotted using MATLAB package considering the different effects of the magnetic flux intensity,thermal radiation parameter,thermal slipping at boundaries,elastic wall properties,and nanoparticle concentration.The results demonstrate the strong impact of the magnetic field and radiation heating in terms of enhancing the nanofluid CSF flow behavior and destroying cancer.
文摘In this work,a computational modelling and analysis framework is developed to investigate the thermal buckling behavior of doubly-curved composite shells reinforced with graphene-origami(G-Ori)auxetic metamaterials.A semi-analytical formulation based on the First-Order Shear Deformation Theory(FSDT)and the principle of virtual displacements is established,and closed-form solutions are derived via Navier’s method for simply supported boundary conditions.The G-Ori metamaterial reinforcements are treated as programmable constructs whose effective thermo-mechanical properties are obtained via micromechanical homogenization and incorporated into the shell model.A comprehensive parametric study examines the influence of folding geometry,dispersion arrangement,reinforcement weight fraction,curvature parameters,and elastic foundation support on the critical buckling temperature(CBT).The results reveal that,under optimal folding geometry and reinforcement alignment with principal stress trajectories,the CBT can increase by more than 150%.Furthermore,the combined effect of G-Ori reinforcement and elastic foundation substantially enhances thermal buckling resistance.These findings establish design guidelines for architected composite shells in applications such as aerospace thermal skins,morphing structures,and thermally-responsive systems,and illustrate the potential of auxetic graphene metamaterials for multifunctional,lightweight,and thermally robust structural components.
文摘Breast cancer screening programs rely heavily on mammography for early detection;however,diagnostic performance is strongly affected by inter-reader variability,breast density,and the limitations of conven-tional computer-aided detection systems.Recent advances in deep learning have enabled more robust and scalable solutions for large-scale screening,yet a systematic comparison of modern object detection architectures on nationally representative datasets remains limited.This study presents a comprehensive quantitative comparison of prominent deep learning–based object detection architectures for Artificial Intelligence-assisted mammography analysis using the MammosighTR dataset,developed within the Turkish National Breast Cancer Screening Program.The dataset comprises 12,740 patient cases collected between 2016 and 2022,annotated with BI-RADS categories,breast density levels,and lesion localization labels.A total of 31 models were evaluated,including One-Stage,Two-Stage,and Transformer-based architectures,under a unified experimental framework at both patient and breast levels.The results demonstrate that Two-Stage architectures consistently outperform One-Stage models,achieving approximately 2%–4%higher Macro F1-Scores and more balanced precision–recall trade-offs,with Double-Head R-CNN and Dynamic R-CNN yielding the highest overall performance(Macro F1≈0.84–0.86).This advantage is primarily attributed to the region proposal mechanism and improved class balance inherent to Two-Stage designs.One-Stage detectors exhibited higher sensitivity and faster inference,reaching Recall values above 0.88,but experienced minor reductions in Precision and overall accuracy(≈1%–2%)compared with Two-Stage models.Among Transformer-based architectures,Deformable DEtection TRansformer demonstrated strong robustness and consistency across datasets,achieving Macro F1-Scores comparable to CNN-based detectors(≈0.83–0.85)while exhibiting minimal performance degradation under distributional shifts.Breast density–based analysis revealed increased misclassification rates in medium-density categories(types B and C),whereas Transformer-based architectures maintained more stable performance in high-density type D tissue.These findings quantitatively confirm that both architectural design and tissue characteristics play a decisive role in diagnostic accuracy.Overall,the study provides a reproducible benchmark and highlights the potential of hybrid approaches that combine the accuracy of Two-Stage detectors with the contextual modeling capability of Transformer architectures for clinically reliable breast cancer screening systems.
文摘Accurate short-term electricity price forecasts are essential for market participants to optimize bidding strategies,hedge risk and plan generation schedules.By leveraging advanced data analytics and machine learning methods,accurate and reliable price forecasts can be achieved.This study forecasts day-ahead prices in Türkiye’s electricity market using eXtreme Gradient Boosting(XGBoost).We benchmark XGBoost against four alternatives—Support Vector Machines(SVM),Long Short-Term Memory(LSTM),Random Forest(RF),and Gradient Boosting(GBM)—using 8760 hourly observations from 2023 provided by Energy Exchange Istanbul(EXIST).All models were trained on an identical chronological 80/20 train–test split,with hyperparameters tuned via 5-fold cross-validation on the training set.XGBoost achieved the best performance(Mean Absolute Error(MAE)=144.8 TRY/MWh,Root Mean Square Error(RMSE)=201.8 TRY/MWh,coefficient of determination(R^(2))=0.923)while training in 94 s.To enhance interpretability and identify key drivers,we employed Shapley Additive Explanations(SHAP),which highlighted a strong association between higher prices and increased natural-gas-based generation.The results provide a clear performance benchmark and practical guidance for selecting forecasting approaches in day-ahead electricity markets.
