Combining machine and deep transfer learning for mediastinal lymph node evaluation in patients with lung cancer

Background The prognosis and survival of patients with lung cancer are likely to deteriorate with metastasis.Using deep-learning in the detection of lymph node metastasis can facilitate the noninvasive calculation of the likelihood of such metastasis,thereby providing clinicians with crucial information to enhance diagnostic precision and ultimately improve patient survival and prognosis.Methods In total,623 eligible patients were recruited from two medical institutions.Seven deep learning models,namely Alex,GoogLeNet,Resnet18,Resnet101,Vgg16,Vgg19,and MobileNetv3(small),were utilized to extract deep image histological features.The dimensionality of the extracted features was then reduced using the Spearman correlation coefficient(r≥0.9)and Least Absolute Shrinkage and Selection Operator.Eleven machine learning methods,namely Support Vector Machine,K-nearest neighbor,Random Forest,Extra Trees,XGBoost,LightGBM,Naive Bayes,AdaBoost,Gradient Boosting Decision Tree,Linear Regression,and Multilayer Perceptron,were employed to construct classification prediction models for the filtered final features.The diagnostic performances of the models were assessed using various metrics,including accuracy,area under the receiver operating characteristic curve,sensitivity,specificity,positive predictive value,and negative predictive value.Calibration and decision-curve analyses were also performed.Results The present study demonstrated that using deep radiomic features extracted from Vgg16,in conjunction with a prediction model constructed via a linear regression algorithm,effectively distinguished the status of mediastinal lymph nodes in patients with lung cancer.The performance of the model was evaluated based on various metrics,including accuracy,area under the receiver operating characteristic curve,sensitivity,specificity,positive predictive value,and negative predictive value,which yielded values of 0.808,0.834,0.851,0.745,0.829,and 0.776,respectively.The validation set of the model was assessed using clinical decision curves,calibration curves,and confusion matrices,which collectively demonstrated the model's stability and accuracy.Conclusion In this study,information on the deep radiomics of Vgg16 was obtained from computed tomography images,and the linear regression method was able to accurately diagnose mediastinal lymph node metastases in patients with lung cancer.

Comprehensive Performance Evaluation of Family Farms from the Perspective of Land Transfer——Taking Rudong County as an Example

With the reform of agriculture, family farms have become hot spots in the reform of agricultural management system. Taking Rudong County of Jiangsu Province for an example, the land transfer performance of family farms was evalu- ated using the first hand data from field investigation by building an index system for comprehensive performance evaluation of family farm land transfer covering three dimensions of economy, society and ecology. The results show that land transfer in family farm mode has realized the scale management of land, enhanced the agri- cultural level of science and technology, effectively promoted the structural adjust- ment of agriculture and income increase of farmers, advanced the development of eco-agriculture, and achieved good economic and social performances. However, in the development process of family farms, inputs of agro-chemicals, fertilizers and other agro-products have brought about the eco-environmental problems, which are still worth noting.

Research on passenger flow control at metro transfer stations based on real-time flow calculation of streamlines

Purpose – The volume of passenger traffic at metro transfer stations serves as a pivotal metric for theorchestration of crowd flow management. Given the intricacies of crowd dynamics within these stations andthe recurrent instances of substantial passenger influxes, a methodology predicated on stochastic processesand the principle of user equilibrium is introduced to facilitate real-time traffic flow estimation within transferstation streamlines.Design/methodology/approach – The synthesis of stochastic process theory with streamline analysisengenders a probabilistic model of intra-station pedestrian traffic dynamics. Leveraging real-time passengerflow data procured from monitoring systems within the transfer station, a gradient descent optimizationtechnique is employed to minimize the cost function, thereby deducing the dynamic distribution of categorizedpassenger flows. Subsequently, adhering to the tenets of user equilibrium, the Frank–Wolfe algorithm isimplemented to allocate the intra-station categorized passenger flows across various streamlines, ascertainingthe traffic volume for each.Findings – Utilizing the Xiaozhai Station of the Xi’an Metro as a case study, the Anylogic simulation softwareis engaged to emulate the intra-station crowd dynamics, thereby substantiating the efficacy of the proposedpassenger flow estimation model. The derived solutions are instrumental in formulating a crowd controlstrategy for Xiaozhai Station during the peak interval from 17:30 to 18:00 on a designated day, yielding crowdmanagement interventions that offer insights for the orchestration of passenger flow and operationalgovernance within metro stations.Originality/value – The construction of an estimation methodology for the real-time streamline traffic flowaugments the model’s dataset, supplanting estimated values derived from surveys or historical datasets withreal-time computed traffic data, thereby enhancing the precision and immediacy of crowd flow managementwithin metro stations.

SERVQUAL Model Based Evaluation Analysis of Railway Passenger Transport Service Quality in China

Railway is the backbone of Chinese transportation system,but its poor quality of services for passengers cause complains now and then.This study first analyzed the influencing factors of service quality on railway passenger,and its quality characteristics was also explained,and finally we proposed an evaluation system of service quality on railway passenger transport.Through the statistical analysis and processing of the basic information from survey data from railway station,trains and the official website of the ticket purchase,the evaluation score of question naire was converted into the score in evaluation index system,which was based on SERVQUAL model.Finally,the evaluation index system was applied to the field test,and all levels of indicators and the overall evaluation of railway passenger transport service quality was obtained.The relevant results show that the evaluation model of this study is concise and practical,and the method has certain practicability and promotion value,which is beneficial to the department of management supervision in railway transportation.

Comprehensive Evaluation and Prediction of Enhancement of Boiling Heat Transfer with Additives

A model of evaluation and prediction of enhancement of boiling heat transfer with additives has been proposed according to fuzzy fundamentals. Correlative appraisement of boiling heat transfer augmentation was done with the model based on 39 additives which were tested by the authors and other researchers. The results show that the evaluation of 35 additives is consistent with experiments, which means that the accuracy of the model is 89.7 percent. In addition, the prediction of the ability of boiling heat transfer enhancement with sodium oleate, polyethylene glycol and Tween-40 is also in good agreement with correspondent experiments.

Performance Evaluation of IEEE 1394b Serial Bus with Deterministic and Stochastic Petri Nets

IEEE 1394b is a high-speed data bus that is widely used in local area networks and consumer electronics. To best utilize this data bus, we need to fully understand its per- formance. To evaluate its performance for both isochronous and asynchronous data transfers, we model the IEEE 1394b bus using Deterministic and Stochastic Pelri Nets （DSPNs）, and conduct simulations using TimeNET 4.0. Using the DSPNs model, we simulate both the average waiting times of isochronous and asynchronous packets. From the simulation results, we determine the factors that influence performance, such as throughput, size of packets, arrival rate of packets, number of isochronous channels, and bus configuration.

Probabilistic interval prediction of metro-to-bus transfer passenger flow in the trip chain

To accurately analyze the fluctuation range of time-varying differences in metro-to-bus transfer passenger flows,the application of a probabilistic interval prediction model is proposed to predict transfer passenger flows.First,bus and metro data are processed and matched by association to construct the basis for public transport trip chain extraction.Second,a reasonable matching threshold method to discriminate the transfer relationship is used to extract the public transport trip chain,and the basic characteristics of the trip based on the trip chain are analyzed to obtain the metro-to-bus transfer passenger flow.Third,to address the problem of low accuracy of point prediction,the DeepAR model is proposed to conduct interval prediction,where the input is the interchange passenger flow,the output is the predicted median and interval of passenger flow,and the prediction scenarios are weekday,non-workday,and weekday morning and evening peaks.Fourth,to reduce the prediction error,a combined particle swarm optimization(PSO)-DeepAR model is constructed using the PSO to optimize the DeepAR model.Finally,data from the Beijing Xizhimen subway station are used for validation,and results show that the PSO-DeepAR model has high prediction accuracy,with a 90%confidence interval coverage of up to 93.6%.

A Sound Quality Objective Evaluation Method Based on Auditory Peripheral Simulation Model

Based on auditory peripheral simulation model, a new Sound Quality Objective Evaluation (SQOE) method is presented,which can be used to model and analyze the impacts of head, shoulder and other parts of human body on sound wave trans-mission.This method employs the artificial head technique, in which the head related transfer function was taken into account tothe outer ear simulation phase.First, a bionic artificial head was designed as the outer ear model with considering the outersound field in view of theory and physical explanations.Then the auditory peripheral simulation model was built, which mimicsthe physiological functions of the human hearing, simulating the acoustic signal transfer process and conversion mechanismsfrom the free field to the peripheral auditory system.Finally, performance comparison was made between the proposed SQOEmethod and ArtemiS software, and the verifications of subjective and objective related analysis were made.Results show thatthe proposed method was economical, simple, and with good evaluation quality.

Analysis of micro vortex metallization and introduction of evaluation factor in mixing process

To improve the mixing efficiency in water purification or wastewater treatment process and understand the mechanism of mass transfer,and a suitable evaluation factor is introduced to evaluate the efficiency of mixing process.The diffusion mass transfer under turbulent condition is investigated based on dynamic and mathematical analysis.The results indicate that submicroscopic diffusion is the rate-limiting step in mass transfer,and the inertia effect of micro vortex causes the phase mixing.An excellent efficiency can be obtained by controlling the scale of micro vortex within millimeter.Furthermore,a new indicator named mixing factor(IH)is proposed to evaluate the efficiency of mixing processes,which is more feasible than conventional evaluation methods,because of its connection of mixing extent with energy consumption.

Animal Exercise:A New Evaluation Method

At present,Animal Exercise courses rely too much on teachers’subjective ideas in teaching methods and test scores,and there is no set of standards as a benchmark for reference.As a result,students guided by different teachers have an uneven understanding of the Animal Exercise and cannot achieve the expected effect of the course.In this regard,the authors propose a scoring system based on action similarity,which enables teachers to guide students more objectively.The authors created QMonkey,a data set based on the body keys of monkeys in the coco dataset format,which contains 1,428 consecutive images from eight videos.The authors use QMonkey to train a model that recognizes monkey body movements.And the authors propose a new non-standing posture normalization method for motion transfer between monkeys and humans.Finally,the authors utilize motion transfer and structural similarity contrast algorithms to provide a reliable evaluation method for animal exercise courses,eliminating the subjective influence of teachers on scoring and providing experience in the combination of artificial intelligence and drama education.

基于云模型的地铁站点与常规公交衔接换乘评价

为了识别地铁站点与常规公交线路衔接换乘存在的换乘效率低等问题,提高城市公共交通系统运行效率和服务水平,促进城市公共交通可持续发展以及响应公众需求和监督,需要对地铁站点与常规公交衔接换乘进行全面系统评价.基于便利性、可靠性、可达性、通畅性及舒适度等5个维度15个指标,建立地铁站点与常规公交的衔接换乘评价指标体系;选取CRITIC(criteria importance through inter-criteria correlation)方法对指标进行客观赋权,采用云模型生成基于基本算法的评价标准云和结果云,评价结果通过云的数字特征和云图直观展现,进而对识别出的问题和不足采取相应改进措施.以中国青岛市主城区45个地铁站点与常规公交衔接换乘为案例进行评价,结果表明,衔接换乘综合评价等级为“中等”,准则层(便利性、可靠性、可达性、通畅性和舒适度)的衔接换乘评价等级均为“中等”以上,基本符合青岛市主城区地铁站点与常规公交衔接换乘的特点,验证了评价模型的有效性和合理性.评价结果有助于优化公共交通资源配置,提高城市交通系统的整体运行效率,缓解城市交通拥堵.

摩托过山车乘客加速度分析与评价

游乐设施乘客加速度是影响乘客生理指标的最重要的技术参数,科学评价乘客加速度是保证游乐设施乘客安全性和乘客束缚装置有效性的重要技术手段。针对摩托过山车乘客加速度分析方法尚不完善的问题,分析并得到了用于评价乘客加速度阈值范围和乘客束缚装置技术参数的测试坐标系及其解析方法并进行了乘客加速度测试。基于5Hz的4阶巴特沃斯低通滤波数据进行了乘客加速度阈值范围评价,结果表明摩托过山车的单向加速度和组合加速度符合标准要求,但YZ组合加速度有超出允许值边界的数据,能够以此为指导对设备进行优化。基于1Hz的4阶巴特沃斯低通滤波数据进行了乘客束缚装置技术参数评价,结果表明摩托过山车乘客束缚装置的技术参数满足加速度区域3所对应的技术要求。通过乘客加速度分析与评价,保证了摩托过山车的乘客安全性和乘客束缚装置有效性。

基于主观评价法的地铁换乘满意度调查——以北京地铁为例

近年来,地铁换乘系数逐年攀升,使得地铁换乘满意度成为衡量换乘效率和乘客体验的综合性指标,对地铁换乘设施的发展至关重要。通过深入剖析换乘环节与乘客心理,构建以换乘速度、换乘便捷度和环境友好度为核心的换乘满意度评价指标体系。为获取乘客对各指标重要性排序与评分,设计网络问卷进行调查。经过数据分析,地铁换乘的规划与建设应将提升便捷性放在首位,包括优化换乘设施布局、完善客流引导系统,并努力缩短换乘的时空距离,营造舒适的换乘环境,以提升乘客换乘满意度。研究对衡量换乘效果、优化换乘站建设具有指导意义,可助力地铁换乘系统的持续优化,并提升城市交通的整体效能与乘客的出行幸福感。

基于Anylogic仿真模型的王府庄站客流组织仿真与优化

为优化轨道交通换乘站客流组织,以济南市轨道交通王府庄站为研究对象,在考虑车站用地规模限制和站内设施布局不变的情况下,利用Anylogic仿真软件构建王府庄站高峰时段客流组织仿真模型,并选用客流密度和平均换乘时间作为客流组织评价指标来分析仿真结果。分析结果发现,王府庄站高峰时段客流存在客流流线交叉、乘客在楼梯走行滞缓和平均换乘时间较长等问题。为此,提出设置导乘标识、分隔栏杆和缩短列车发车间隔等优化方案,并通过优化前后的仿真数据进行对比检验。结果表明:同时采用3种优化方法后,高峰时段王府庄站内的客流交叉得到缓解,1号线站台楼梯的客流密度下降34.3%,2号线站台楼梯的客流密度下降37.5%,站内乘客平均换乘时间下降25%。

Risk evaluation for the task transfer of an aircraft maintenance program based on a multielement connection number

This paper proposes a framework for evaluating the efficacy and suitability of maintenance programs with a focus on quantitative risk assessment in the domain of aircraft maintenance task transfer. The analysis is anchored in the principles of Maintenance Steering Group-3 (MSG-3) logic decision paradigms. The paper advances a holistic risk assessment index architecture tailored for the task transfer of maintenance programs. Utilizing the analytic network process (ANP), the study quantifies the weight interrelationships among diverse variables, incorporating expert-elicited subjective weighting. A multielement connection number-based evaluative model is employed to characterize decision-specific data, thereby facilitating the quantification of task transfer-associated risk through the appraisal of set-pair potentials. Moreover, the paper conducts a temporal risk trend analysis founded on partial connection numbers of varying orders. This analytical construct serves to streamline the process of risk assessment pertinent to maintenance program task transfer. The empirical component of this research, exemplified through a case study of the Boeing 737NG aircraft maintenance program, corroborates the methodological robustness and pragmatic applicability of the proposed framework in the quantification and analysis of mission transfer risk.

地方科研院所科技成果转化绩效评价体系构建

分析地方科研院所在科技成果转化领域的现状与面临的挑战,阐述构建绩效评价体系的迫切需求。调查结果显示,地方科研院所在转化科技成果时,普遍存在科研成果与市场需求不匹配、转化人才供给不足、技术成熟度评估准确性欠佳以及资本运作机制薄弱等难题。为应对这些挑战,设计一套综合性的绩效评价体系,覆盖了技术成熟度、市场适应性、经济效益、社会影响以及创新驱动等关键维度。通过应用实例验证,该体系展现出了其在提升科技成果转化效率与质量、优化资源配置、激励科研人员积极参与转化活动以及促进区域经济与创新力提升方面的显著效用。

Experimental research on stabilities, thermophysical properties and heat transfer enhancement of nanofluids in heat exchanger systems

Stable TiO_2–water nanofluids are prepared by a two-step method, stabilities of nanofluids are investigated by precipitation method and transmittance method respectively, and thermal conductivities and viscosities are also measured. An experimental system for studying the heat transfer enhancement of nanofluids is established,and heat transfer and flow characteristics of TiO_2–water nanofluids in heat exchanger systems with a triangular tube and circular tube are experimentally studied. The effects of nanoparticle mass fractions(ω = 0.1 wt%–0.5 wt%) and Reynolds numbers(Re = 800–10000) on the heat transfer and flow performances of nanofluids are analyzed. Fitting formulas for Nusselt number and resistance coefficient of nanofluids in a triangular tube are put forward based on the experimental data. The comprehensive performances of nanofluids in a triangular tube are investigated. It is found that nanofluids in a triangular tube can significantly improve the heat transfer performance at the cost of a small increase in resistance coefficient compared with that in a circular tube, especially the resistance coefficients are almost the same between different nanoparticle mass fractions at turbulent flow. It is also found that the comprehensive evaluation index η decreases with Reynolds number at laminar flow but a critical maximum value appears at turbulent flow.

Comparative Study of Transfer Learning Models for Retinal Disease Diagnosis from Fundus Images

While the usage of digital ocular fundus image has been widespread in ophthalmology practice,the interpretation of the image has been still on the hands of the ophthalmologists which are quite costly.We explored a robust deep learning system that detects three major ocular diseases:diabetic retinopathy(DR),glaucoma(GLC),and age-related macular degeneration(AMD).The proposed method is composed of two steps.First,an initial quality evaluation in the classification system is proposed to filter out poorquality images to enhance its performance,a technique that has not been explored previously.Second,the transfer learning technique is used with various convolutional neural networks(CNN)models that automatically learn a thousand features in the digital retinal image,and are based on those features for diagnosing eye diseases.Comparison performance of many models is conducted to find the optimal model which fits with fundus classification.Among the different CNN models,DenseNet-201 outperforms others with an area under the receiver operating characteristic curve of 0.99.Furthermore,the corresponding specificities for healthy,DR,GLC,andAMDpatients are found to be 89.52%,96.69%,89.58%,and 100%,respectively.These results demonstrate that the proposed method can reduce the time-consumption by automatically diagnosing multiple eye diseases using computer-aided assistance tools.

Heat Transfer Performance of a Novel Microchannel Embedded with Connected Grooves

To improve the heat transfer performance of microchannels,a novel microchannel embedded with connected grooves crossing two sidewalls and the bottom surface(type A)was designed.A comparative study of heat transfer was conducted regarding the performances of type A microchannels,microchannels embedded with grooves on their bottom(including types B and C),or on the sidewalls(type D)as well as smooth rectangular microchannels(type E)via a three-dimensional numerical simulation and experimental validation(at Reynolds numbers from 118 to 430).Numerical results suggested that the average Nusselt number of types A,B,C,and D microchannels were 106,73.4,50.1,and 12.6%higher than that of type E microchannel,respectively.The smallest synergy angle β and entropy generation number Ns,a were determined for type A microchannels based on field synergy and nondimensional entropy analysis,which indicated that type A exhibited the best heat transfer performance.Numerical flow analysis indicated that connected grooves induced fluid to flow along two different temperature gradients,which contributed to enhanced heat transfer performance.

Study on crowd evacuation in subway transfer station fires based on numerical simulation

In order to study the influencing factors of fire evacuation bottlenecks and evacuation time in subway transfer stations,a subway transfer station model was built using Pathfinder to simulate the emergency evacuation of passengers in a fire.Using the control variable method,the evacuation under different conditions is simulated by changing the parameters.The effects of the use of escalators,the speed of escalators,preference for stairs and escalators,the use of escalators,the familiarity of exits,the speed of personnel movement and the width of stairs on evacuation time are discussed and analyzed.The results show that the fire evacuation bottlenecks of subway transfer stations is each stair entrance.Evacuation time can be shortened by increasing the speed of escalators,increasing the proportion of escalator walkers,reducing the proportion of passengers who choose familiar exits to escape,increasing the speed of passengers and increasing the width of stairs.