Using Cross Entropy as a Performance Metric for Quantifying Uncertainty in DNN Image Classifiers: An Application to Classification of Lung Cancer on CT Images

Cross entropy is a measure in machine learning and deep learning that assesses the difference between predicted and actual probability distributions. In this study, we propose cross entropy as a performance evaluation metric for image classifier models and apply it to the CT image classification of lung cancer. A convolutional neural network is employed as the deep neural network (DNN) image classifier, with the residual network (ResNet) 50 chosen as the DNN archi-tecture. The image data used comprise a lung CT image set. Two classification models are built from datasets with varying amounts of data, and lung cancer is categorized into four classes using 10-fold cross-validation. Furthermore, we employ t-distributed stochastic neighbor embedding to visually explain the data distribution after classification. Experimental results demonstrate that cross en-tropy is a highly useful metric for evaluating the reliability of image classifier models. It is noted that for a more comprehensive evaluation of model perfor-mance, combining with other evaluation metrics is considered essential. .

Effect of Using Israel Napier Grass in Feed Ration on Growth Performance of Local Chicken of Cambodia

This study investigated the effect of Israel grass in feed formulation on growth performance of local chicken.The study was conducted from October 2023 until January 2024 in Kampong Chheutel Institute of Technology.The CRD(Completely Randomized Design)was used with 4 treatments and 4 replications.The local chicks were purchased from local famer at 1 day old and kept in whole pen by feeding commercial feed.At 21 days old the total 320 chicks were allocated into experimental treatments.The 4 feed formulas based on the level of Israel grass added such as 0,5%,7%and 9%,were called T1(control),T2,T3 and T4 respectively.The result showed that body weight gain,feed intake,ADG(Average Daily Gain)and FCR(Feed Conversion Ratio)in all treatments were not significantly different(p>0.05).However,there was strong correlation between living weight with carcass of al treatments,but no difference among carcass percentage of local chicken for all different diet in this study.In conclusion,the Israel grass did not affect growth performance of local chickens and can be used to replace some expensive feed ingredient,such as rice bran.

Enhancing performance of low-temperature processed CsPbI2Br all-inorganic perovskite solar cells using polyethylene oxide-modified TiO_(2)

CsPbX_(3)-based(X=I,Br,Cl)inorganic perovskite solar cells(PSCs)prepared by low-temperature process have attracted much attention because of their low cost and excellent thermal stability.However,the high trap state density and serious charge recombination between low-temperature processed TiO_(2)film and inorganic perovskite layer interface seriously restrict the performance of all-inorganic PSCs.Here a thin polyethylene oxide(PEO)layer is employed to modify TiO_(2)film to passivate traps and promote carrier collection.The impacts of PEO layer on microstructure and photoelectric characteristics of TiO_(2)film and related devices are systematically studied.Characterization results suggest that PEO modification can reduce the surface roughness of TiO_(2)film,decrease its average surface potential,and passivate trap states.At optimal conditions,the champion efficiency of CsPbI_(2)Br PSCs with PEO-modified TiO_(2)(PEO-PSCs)has been improved to 11.24%from 9.03%of reference PSCs.Moreover,the hysteresis behavior and charge recombination have been suppressed in PEO-PSCs.

Effects of thinning on the understory light environment of different stands and the photosynthetic performance and growth of the reforestation species Phoebe bournei

Light levels determine regeneration in stands and a key concern is how to regulate the light environment of different stand types to the requirements of the understory.In this study,we selected three stands typical in south China(a Cryptomeria japonica plantation,a Quercus acutissima plantation,and a mixed stand of both)and three thinning intensities to determine the best understory light environ-ment for 3-year-old Phoebe bournei seedlings.The canopy structure,understory light environment,and photosynthe-sis and growth indicators were assessed following thin-ning.Thinning improved canopy structure and understory light availability of each stand;species composition was the reason for differences in the understory light environ-ment.Under the same thinning intensity,the mixed stand had the greatest light radiation and most balanced spectral composition.P.bournei photosynthesis and growth were closely related to the light environment;all three stands required heavy thinning to create an effective and sustained understory light environment.In a suitable understory light environment,the efficiency of light interception,absorption,and use by seedlings was enhanced,resulting in a higher carbon assimilation the main limiting factor was stomatal conductance.As a shade-avoidance signal,red/far-red radia-tion is a critical factor driving changes in photosynthesis and growth of P.bournei seedlings,and a reduction increased light absorption and use capacity and height:diameter ratios.The growth advantage transformed from diameter to height,enabling seedlings to access more light.Our findings suggest that the regeneration of shade-tolerant species such as P.bournei could be enhanced if a targeted approach to thinning based on stand type was adopted.

Semi-hydro-equivalent design and performance extrapolation between 100 kJ-scale and NIF-scale indirect drive implosion

Extrapolation of implosion performance between different laser energy scales is investigated for indirect drive through a semi-hydroequivalent design.Since radiation transport is non-hydro-equivalent,the peak radiation temperature of the hohlraum and the ablation velocity of the capsule ablator are not scale-invariant when the sizes of the hohlraum and the capsule are scale-varied.A semi-hydro-equivalent design method that keeps the implosion velocity V_(i),adiabat α_(F),and P_(L)/R_(hc)^(2) (where P_(L) is the laser power and R_(hc) is the hohlraum and capsule scale length)scale-invariant,is proposed to create hydrodynamically similar implosions.The semi-hydro-equivalent design and the scaled implosion performance are investigated for the 100 kJ Laser Facility(100 kJ-scale)and the National Ignition Facility(NIF-scale)with about 2 MJ laser energy.It is found that the one-dimensional implosion performance is approximately hydro-equivalent when V_(i) and α_(F) are kept the same.Owing to the non-hydro-equivalent radiation transport,the yield-over-clean without α-particle heating(YOC_(noα))is slightly lower at 100 kJ-scale than at NIF-scale for the same scaled radiation asymmetry or the same initial perturbation of the hydrodynamic instability.The overall scaled two-dimensional implosion performance is slightly lower at 100 kJ-scale.The general Lawson criterion factor scales asχ_(noα) ^(2D)∼S^(1.06±0.04)(where S is the scale-variation factor)for the semi-hydro-equivalent implosion design with a moderate YOC_(noα).Our study indicates that χ_(noα)≈0.379 is the minimum requirement for the 100 kJ-scale implosion to demonstrate the ability to achieve marginal ignition at NIF-scale.

Two-Way Neural Network Performance PredictionModel Based onKnowledge Evolution and Individual Similarity

Predicting students’academic achievements is an essential issue in education,which can benefit many stakeholders,for instance,students,teachers,managers,etc.Compared with online courses such asMOOCs,students’academicrelateddata in the face-to-face physical teaching environment is usually sparsity,and the sample size is relativelysmall.It makes building models to predict students’performance accurately in such an environment even morechallenging.This paper proposes a Two-WayNeuralNetwork(TWNN)model based on the bidirectional recurrentneural network and graph neural network to predict students’next semester’s course performance using only theirprevious course achievements.Extensive experiments on a real dataset show that our model performs better thanthe baselines in many indicators.

Multi-Model Approach for Assessing the Influence of Calibration Criteria on the Water Balance in Ouémé Basin

Hydrological models are very useful tools for evaluating water resources, and the hydroclimatic hazards associated with the water cycle. However, their calibration and validation require the use of performance criteria which choice is not straightforward. This paper aims to evaluate the influence of the performance criteria on water balance components and water extremes using two global rainfall-runoff models (HBV and GR4J) over the Ouémé watershed at the Bonou and Savè outlets. Three (3) Efficacy criteria (Nash, coefficient of determination, and KGE) were considered for calibration and validation. The results show that the Nash criterion provides a good assessment of the simulation of the different parts of the hydrograph. KGE is better for simulating peak flows and water balance elements than other efficiency criteria. This study could serve as a basis for the choice of performance criteria in hydrological modelling.

Elliptical encirclement control capable of reinforcing performances for UAVs around a dynamic target

Most researches associated with target encircling control are focused on moving along a circular orbit under an ideal environment free from external disturbances.However,elliptical encirclement with a time-varying observation radius,may permit a more flexible and high-efficacy enclosing solution,whilst the non-orthogonal property between axial and tangential speed components,non-ignorable environmental perturbations,and strict assignment requirements empower elliptical encircling control to be more challenging,and the relevant investigations are still open.Following this line,an appointed-time elliptical encircling control rule capable of reinforcing circumnavigation performances is developed to enable Unmanned Aerial Vehicles(UAVs)to move along a specified elliptical path within a predetermined reaching time.The remarkable merits of the designed strategy are that the relative distance controlling error can be guaranteed to evolve within specified regions with a designer-specified convergence behavior.Meanwhile,wind perturbations can be online counteracted based on an unknown system dynamics estimator(USDE)with only one regulating parameter and high computational efficiency.Lyapunov tool demonstrates that all involved error variables are ultimately limited,and simulations are implemented to confirm the usability of the suggested control algorithm.

Dietary supplementation of laminarin improves the reproductive performance of sows and the growth of suckling piglets

Background Maternal nutrition is essential in keeping a highly efficient production system in the pig industry.Laminarin has been shown to improve antioxidant capacity,reduce the inflammatory response,and favor the homeostasis of intestinal microbiota.However,the effect of dietary supplementation of laminarin on the reproductive performance of sows and the growth of suckling offspring remains unknown.Methods A total of 40 Landrace×Yorkshire multiparous sows on d 85 of gestation,similar in age,body weight(BW),parity and reproductive performance,were randomly divided into four dietary treatments with 10 sows per treatment,receiving a control diet(basal pregnancy or lactating diets)and a basal diet supplemented with 0.025%,0.05%and 0.10%laminarin,respectively.The experiment lasted from d 85 of gestation to d 21 of lactation.Results Laminarin supplementation linearly increased number born alive per litter(P=0.03),average daily feed intake(ADFI,P<0.01),and total milk yield of sows during the lactation of 1–21 d(P=0.02).Furthermore,maternal laminarin supplementation increased the average daily gain(ADG)of piglets while tending to reduce the culling and death rate before weaning.In addition,alterations to the composition of colostrum and milk,as well as to serum inflammatory cytokines and immunoglobulins of sows were observed.The fecal microbiota profile of sows supported the improvement of reproductive performance in sows and the growth performance in suckling offspring.Conclusions Dietary supplementation of laminarin during late pregnancy and lactation could significantly improve reproductive performance of sows and growth performance of piglets.

Effect of Molecular Weight on Thermoelectric Performance of P3HT Analogues with 2-Propoxyethyl Side Chains

By replacing hexyl chains in poly(3-hexylthiophene)(P3HT)with 2-propoxyethyls,four poly(3-(2-propoxyethyl)thiophene)(P3POET)homopolymers with comparable polydispersity indexes(PDI)and regioregularities were prepared herein in addition with step increment of about 7 kDa on numberaverage molecular weight(M_(n))from around 11 to 32 kDa(accordingly denoted as P11k,P18k,P25k,and P32k).When doped in film by FeCl_(3)at the optimized conditions,the maximum power factor(PF_(max))increases greatly from 4.3μW·m^(-1)·K^(-2)for P11k to 8.8μW·m^(-1)·K^(-2)for P18k,and further to 9.7μW·m^(-1)·K^(-2)for P25k,followed by a slight decrease to 9.2μW·m^(-1)·K^(-2)for P32k.The close Seebeck coefficients(S)at PF_(max)are observed in these doped polymer films due to their consistent frontier orbital energy levels and Fermi levels.The main contribution to this PF_(max)evolution thus comes from the corresponding conductivities(σ).Theσvariation of the doped films can be rationally correlated with their microstructure evolution.

Increased number of symptoms during the acute phase of SARS-CoV-2 infection in athletes is associated with prolonged time to return to full sports performance—AWAREⅧ

Purpose The aim of the study was to identify factors associated with prolonged time to return to full performance(RTFP)in athletes with recent severe acute respiratory syndrome coronavirus 2(SARS-CoV-2)infection.Methods Prospective cohort study with cross sectional analysis.A total of 84 athletes with confirmed SARS-CoV-2 infection assessed at a coronavirus disease 2019 recovery clinic gave a history of age,sex,type/level of sport,co-morbidities,pre-infection training hours,and 26 acute SARS-CoV-2 symptoms from 3 categories(“nose and throat”,“chest and neck”,and“whole body”/systemic).Data on days to RTFP were obtained by structured interviews.Factors associated with RTFP were demographics,sport participation,history of co-morbidities,pre-infection training history,and acute symptoms(type,number).Outcomes were:(a)days to RTFP(median,interquartile range(IQR))in asymptomatic(n=7)and symptomatic athletes(n=77),and(b)hazard ratios(HRs;95%confidence interval)for symptomatic athletes with vs.without a factor(univariate,multiple models).HR<1 was predictive of higher percentage chance of prolonged RTFP.Significance was p<0.05.Results Days to RTFP were 30 days(IQR:23–40)for asymptomatic and 64 days(IQR:42–91)for symptomatic participants(p>0.05).Factors associated with prolonged RTFP(univariate models)were:females(HR=0.57;p=0.014),endurance athletes(HR=0.41;p<0.0001),co-morbidity number(HR=0.75;p=0.001),and respiratory disease history(HR=0.54;p=0.026).In symptomatic athletes,prolonged RTFP(multiple models)was significantly associated with increased“chest and neck”(HR=0.85;p=0.017)and“nose and throat”(HR=0.84;p=0.013)symptoms,but the association was more profound between prolonged RFTP and increased total number of“all symptoms”(HR=0.91;p=0.001)and“whole body”/systemic(HR=0.82;p=0.007)symptoms.Conclusion A larger number of total symptoms and specifically“whole body”/systemic symptoms during the acute phase of SARS-CoV-2 infection in athletes is associated with prolonged RTFP.

Effects of Additives on the Microstructure and Tribology Performance of Ta-12W Alloy Micro-Arc Oxidation Coating

Oxide ceramic coatings were fabricated on tantalum alloys by micro-arc oxidation (MAO) to improve their hardness and tribological properties. The MAO coatings were manufactured in a mixed silicatephosphate electrolyte containing NaF and/or EDTA (ethylene diamine tetraacetic acid). The surface morphology,cross-sectional view, chemical composition, hardness, and wear performance of the coatings were analysed. As revealed by the scanning electron microscopy, silica-rich nodules appear on the MAO coating obtained in the silicate-phosphate electrolyte, but the formation of nodules is inhibited with NaF and/or EDTA in the electrolyte.Also, they reduce the roughness and improve the compactness of the coatings, which are composed of Ta_(2)O_(5),(Ta, O), and TaO. A thick and hard coating is obtained in the NaF-containing electrolyte, and the tribology performance is effectively improved. With additives, the nodule structure is detached from the coating surface and dissolved in the electrolyte. By using NaF as an electrolyte additive, the abrasion performance of the MAO coating is enhanced by decreasing the nodule structure, increasing the size of micropores, and improving the coating hardness.

Optimal investment based on relative performance and weighted utility

This paper studies the optimal portfolio allocation of a fund manager when he bases decisions on both the absolute level of terminal relative performance and the change value of terminal relative performance comparison to a prede ned reference point.We nd the optimal investment strategy by maximizing a weighted average utility of a concave utility and an S-shaped utility via a concavi cation technique and the martingale method.Numerical results are carried out to show the impact of the extent to which the manager pays attention to the change of relative performance related to the reference point on the optimal terminal relative performance.

The Impact of Health Information Technology on Hospital Performance: A Systematic Integrative Literature Review

Objective: To review, categorise, and synthesise findings from literature on health information technology (HIT) functionalities, HIT use, and the impact of HIT on hospital performance. Materials and Methods: We conducted a systematic integrative literature review based on a compre-hensive database search. To organise, categorise and synthesise the ex-isting literature, we adopted the affordance actualization theory. To align the literature with our research framework, we used four categories: 1) the functionalities of HIT and how these functionalities are measured;2) use and immediate outcomes of HIT functionalities;3) different perfor-mance indicators and how HIT functionalities affect them;and 4) what hospital characteristics influence the outcome of hospital performance. Results: Fifty-two studies were included. We identified four types of HIT. Only ten studies (19.2%) define the use of HIT by explicitly meas-uring the use rate of HIT. We identified five dimensions of hospital per-formance indicators. Every dimension showed mixed results;however, in general, HIT has a positive impact on mortality and patient readmis-sions. We found several hospital characteristics that may affect the rela-tionship between HIT and hospital-level outcomes. Discussion: Further efforts should focus on embedded research on HIT functionalities, use and effects of HIT implementations with more performance indicators and adjusted for hospital characteristics. Conclusion: The proposed framework could help hospitals and researchers make decisions regard-ing the functionalities, use and effects of HIT implementation in hospi-tals. Given our research outcomes, we suggest future research opportuni-ties to improve understanding of how HIT affects hospital performance. .

Adaptive admittance tracking control for interactive robot with prescribed performance

An adaptive control approach is presented in this paper for tracking desired trajectories in interactive manipulators. The controller design incorporates prescribed performance functions (PPFs) to improve dynamic performance. Notably, the performance of the output error is confined in an envelope characterized by exponential convergence, leading to convergence to zero. This feature ensures a prompt response from admittance control and establishes a reliable safety framework for interactions. Simulation results provide practical insights,demonstrating the viability of the control scheme proposed in this paper.

Ballistic performance of additive manufacturing 316l stainless steel projectiles based on topology optimization method

Material and structure made by additive manufacturing(AM)have received much attention lately due to their flexibility and ability to customize complex structures.This study first implements multiple objective topology optimization simulations based on a projectile perforation model,and a new topologic projectile is obtained.Then two types of 316L stainless steel projectiles(the solid and the topology)are printed in a selective laser melt(SLM)machine to evaluate the penetration performance of the projectiles by the ballistic test.The experiment results show that the dimensionless specific kinetic energy value of topologic projectiles is higher than that of solid projectiles,indicating the better penetration ability of the topologic projectiles.Finally,microscopic studies(scanning electron microscope and X-ray micro-CT)are performed on the remaining projectiles to investigate the failure mechanism of the internal structure of the topologic projectiles.An explicit dynamics simulation was also performed,and the failure locations of the residual topologic projectiles were in good agreement with the experimental results,which can better guide the design of new projectiles combining AM and topology optimization in the future.

The component-activity interrelationship of cobalt-based bifunctional electrocatalysts for overall water splitting:Strategies and performance

Cobalt-based electrocatalysts take advantage of potentially harmonizable microstructure and flexible coupling effects compared to commercial noble metal-based catalytic materials.However,conventional water electrolysis systems based on cobalt-based monofunctional hydrogen evolution reaction(HER)or oxygen evolution reaction(OER)catalysts have certain shortcomings in terms of resource utilization and universality.In contrast,cobalt-based bifunctional catalysts(CBCs)have attracted much attention in recent years for overall water splitting systems because of their practicality and reduced preparation cost of electrolyzer.This review aims to address the latest development in CBCs for total hydrolysis.The main modification strategies of CBCs are systematically classified in water electrolysis to provide an overview of how to regulate their morphology and electronic configuration.Then,the catalytic performance of CBCs in total-hydrolysis is summarized according to the types of cobalt-based phosphides,sulfides and oxides,and the mechanism of strengthened electrocatalytic ability is emphasized through combining experiments and theoretical calculations.Future efforts are finally suggested to focus on exploring the dynamic conversion of reaction intermediates and building near-industrial CBCs,designing advanced CBC materials through micro-modulation,and addressing commercial applications.

Performance Comparison of Hyper-V and KVM for Cryptographic Tasks in Cloud Computing

As the extensive use of cloud computing raises questions about the security of any personal data stored there,cryptography is being used more frequently as a security tool to protect data confidentiality and privacy in the cloud environment.A hypervisor is a virtualization software used in cloud hosting to divide and allocate resources on various pieces of hardware.The choice of hypervisor can significantly impact the performance of cryptographic operations in the cloud environment.An important issue that must be carefully examined is that no hypervisor is completely superior in terms of performance;Each hypervisor should be examined to meet specific needs.The main objective of this study is to provide accurate results to compare the performance of Hyper-V and Kernel-based Virtual Machine(KVM)while implementing different cryptographic algorithms to guide cloud service providers and end users in choosing the most suitable hypervisor for their cryptographic needs.This study evaluated the efficiency of two hypervisors,Hyper-V and KVM,in implementing six cryptographic algorithms:Rivest,Shamir,Adleman(RSA),Advanced Encryption Standard(AES),Triple Data Encryption Standard(TripleDES),Carlisle Adams and Stafford Tavares(CAST-128),BLOWFISH,and TwoFish.The study’s findings show that KVM outperforms Hyper-V,with 12.2%less Central Processing Unit(CPU)use and 12.95%less time overall for encryption and decryption operations with various file sizes.The study’s findings emphasize how crucial it is to pick a hypervisor that is appropriate for cryptographic needs in a cloud environment,which could assist both cloud service providers and end users.Future research may focus more on how various hypervisors perform while handling cryptographic workloads.

Anti-aging performance improvement and enhanced combustion efficiency of boron via the coating of PDA

Boron is an ambitious fuel in energetic materials since its high heat release values,but its application is prohibited by low combustion efficiency and oxidization during storage.The polydopamine(PDA)was introduced into boron particles,investigating the impact of PDA content on the energetic behavior of boron.The results indicated that the PDA coating formed a fishing net structure on the surface of boron particles.The heat release results showed that the combustion calorific value of B@PDA was higher than that of the raw boron.Specifically,the actual combustion heat of boron powder in B@10%PDA increased by 38.08%.Meanwhile,the DSC peak temperature decreased by 100.65℃under similar oxidation rate compared to raw boron.Simultaneously,the B@PDA@AP and B@AP composites were prepared,and their combustion properties were evaluated.It was demonstrated that B@10%PDA@AP exhibited superior performance in terms of peak pressure and burning time,respectively.The peak pressure is 12.43 kPa more than B@AP and burning time is 2.22 times higher than B@AP.Therefore,the coating of PDA effectively inhibits the oxidization of boron during storage and enhances the energetic behavior of boron and corresponding composites.

A PID Tuning Approach for Inertial Systems Performance Optimization

In the practice of control the industrial processes, proportional-integral-derivative controller remains pivotal due to its simple structure and system performance-oriented tuning process. In this paper are presented two approaches for synthesis the proportional-integral-derivative controller to the models of objects with inertia, that offer the procedure of system performance optimization based on maximum stability degree criterion. The proposed algorithms of system performance optimization were elaborated for model of objects with inertia second and third order and offer simple analytical expressions for tuning the PID controller. Validation and verification are conducted through computer simulations using MATLAB, demonstrating successful performance optimization and showcasing the effectiveness PID controllers’ tuning. The proposed approaches contribute insights to the field of control, offering a pathway for optimizing the performance of second and third-order inertial systems through robust controller synthesis.