Graph Federated Learning(GFL)has shown great potential in privacy protection and distributed intelligence through distributed collaborative training of graph-structured data without sharing raw information.However,exi...Graph Federated Learning(GFL)has shown great potential in privacy protection and distributed intelligence through distributed collaborative training of graph-structured data without sharing raw information.However,existing GFL approaches often lack the capability for comprehensive feature extraction and adaptive optimization,particularly in non-independent and identically distributed(NON-IID)scenarios where balancing global structural understanding and local node-level detail remains a challenge.To this end,this paper proposes a novel framework called GFL-SAR(Graph Federated Collaborative Learning Framework Based on Structural Amplification and Attention Refinement),which enhances the representation learning capability of graph data through a dual-branch collaborative design.Specifically,we propose the Structural Insight Amplifier(SIA),which utilizes an improved Graph Convolutional Network(GCN)to strengthen structural awareness and improve modeling of topological patterns.In parallel,we propose the Attentive Relational Refiner(ARR),which employs an enhanced Graph Attention Network(GAT)to perform fine-grained modeling of node relationships and neighborhood features,thereby improving the expressiveness of local interactions and preserving critical contextual information.GFL-SAR effectively integrates multi-scale features from every branch via feature fusion and federated optimization,thereby addressing existing GFL limitations in structural modeling and feature representation.Experiments on standard benchmark datasets including Cora,Citeseer,Polblogs,and Cora_ML demonstrate that GFL-SAR achieves superior performance in classification accuracy,convergence speed,and robustness compared to existing methods,confirming its effectiveness and generalizability in GFL tasks.展开更多
To better understand the migration behavior of plastic fragments in the environment,development of rapid non-destructive methods for in-situ identification and characterization of plastic fragments is necessary.Howeve...To better understand the migration behavior of plastic fragments in the environment,development of rapid non-destructive methods for in-situ identification and characterization of plastic fragments is necessary.However,most of the studies had focused only on colored plastic fragments,ignoring colorless plastic fragments and the effects of different environmental media(backgrounds),thus underestimating their abundance.To address this issue,the present study used near-infrared spectroscopy to compare the identification of colored and colorless plastic fragments based on partial least squares-discriminant analysis(PLS-DA),extreme gradient boost,support vector machine and random forest classifier.The effects of polymer color,type,thickness,and background on the plastic fragments classification were evaluated.PLS-DA presented the best and most stable outcome,with higher robustness and lower misclassification rate.All models frequently misinterpreted colorless plastic fragments and its background when the fragment thickness was less than 0.1mm.A two-stage modeling method,which first distinguishes the plastic types and then identifies colorless plastic fragments that had been misclassified as background,was proposed.The method presented an accuracy higher than 99%in different backgrounds.In summary,this study developed a novel method for rapid and synchronous identification of colored and colorless plastic fragments under complex environmental backgrounds.展开更多
Although conventional object detection methods achieve high accuracy through extensively annotated datasets,acquiring such large-scale labeled data remains challenging and cost-prohibitive in numerous real-world appli...Although conventional object detection methods achieve high accuracy through extensively annotated datasets,acquiring such large-scale labeled data remains challenging and cost-prohibitive in numerous real-world applications.Few-shot object detection presents a new research idea that aims to localize and classify objects in images using only limited annotated examples.However,the inherent challenge in few-shot object detection lies in the insufficient sample diversity to fully characterize the sample feature distribution,which consequently impacts model performance.Inspired by contrastive learning principles,we propose an Implicit Feature Contrastive Learning(IFCL)module to address this limitation and augment feature diversity for more robust representational learning.This module generates augmented support sample features in a mixed feature space and implicitly contrasts them with query Region of Interest(RoI)features.This approach facilitates more comprehensive learning of both intra-class feature similarity and inter-class feature diversity,thereby enhancing the model’s object classification and localization capabilities.Extensive experiments on PASCAL VOC show that our method achieves a respective improvement of 3.2%,1.8%,and 2.3%on 10-shot of three Novel Sets compared to the baseline model FPD.展开更多
This study utilized a sequential mediating model to examine the role of motivation to learn and transfer selfefficacy in the relationships between perceived content validity,mentoring function,continuous learning work...This study utilized a sequential mediating model to examine the role of motivation to learn and transfer selfefficacy in the relationships between perceived content validity,mentoring function,continuous learning work culture and intention to transfer learning.The sample comprized 429 final-year apprentices in Guangdong province,China(females=69.9%,Engineering&Medicine=69%,mean age=20.99,SD=1.60).The apprentices completed standardized measures of motivation to learn,transfer self-efficacy perceived content validity,mentoring function,and continuous learning work culture.Structural equation modeling was used to analyze the data.Results showed perceived content validity,mentoring function,continuous learning culture to predict intention to transfer learning.Of these factors,perceived content validity was the strongest predictor of intention to transfer learning.Of these factors,perceived content validity was the most influential predictor of intention to transfer learning.The motivation to learn and transfer self-efficacy sequentially mediated the relationship between mentoring function and intention to learning transfer to be stronger than by either alone.Although perceived content validity and continuous learning culture exhibited no significant direct effects on intention to transfer learning,they demonstrated positive indirect associations with intention to transfer via motivation to learn and transfer self-efficacy.These study findings extend the applications of the learning transfer framework to individuals undergoing apprenticeship training which also would apply to other a long-term work-based learning programs.展开更多
Hepatocellular carcinoma(HCC)remains a leading cause of cancer-related mortality globally,necessitating advanced diagnostic tools to improve early detection and personalized targeted therapy.This review synthesizes ev...Hepatocellular carcinoma(HCC)remains a leading cause of cancer-related mortality globally,necessitating advanced diagnostic tools to improve early detection and personalized targeted therapy.This review synthesizes evidence on explainable ensemble learning approaches for HCC classification,emphasizing their integration with clinical workflows and multi-omics data.A systematic analysis[including datasets such as The Cancer Genome Atlas,Gene Expression Omnibus,and the Surveillance,Epidemiology,and End Results(SEER)datasets]revealed that explainable ensemble learning models achieve high diagnostic accuracy by combining clinical features,serum biomarkers such as alpha-fetoprotein,imaging features such as computed tomography and magnetic resonance imaging,and genomic data.For instance,SHapley Additive exPlanations(SHAP)-based random forests trained on NCBI GSE14520 microarray data(n=445)achieved 96.53%accuracy,while stacking ensembles applied to the SEER program data(n=1897)demonstrated an area under the receiver operating characteristic curve of 0.779 for mortality prediction.Despite promising results,challenges persist,including the computational costs of SHAP and local interpretable model-agnostic explanations analyses(e.g.,TreeSHAP requiring distributed computing for metabolomics datasets)and dataset biases(e.g.,SEER’s Western population dominance limiting generalizability).Future research must address inter-cohort heterogeneity,standardize explainability metrics,and prioritize lightweight surrogate models for resource-limited settings.This review presents the potential of explainable ensemble learning frameworks to bridge the gap between predictive accuracy and clinical interpretability,though rigorous validation in independent,multi-center cohorts is critical for real-world deployment.展开更多
Indoor scene semantic segmentation is essential for enabling robots to understand and interact with their environments effectively.However,numerous challenges remain unresolved,particularly in single-robot systems,whi...Indoor scene semantic segmentation is essential for enabling robots to understand and interact with their environments effectively.However,numerous challenges remain unresolved,particularly in single-robot systems,which often struggle with the complexity and variability of indoor scenes.To address these limitations,we introduce a novel multi-robot collaborative framework based on multiplex interactive learning(MPIL)in which each robot specialises in a distinct visual task within a unified multitask architecture.During training,the framework employs task-specific decoders and cross-task feature sharing to enhance collaborative optimisation.At inference time,robots operate independently with optimised models,enabling scalable,asynchronous and efficient deployment in real-world scenarios.Specifically,MPIL employs specially designed modules that integrate RGB and depth data,refine feature representations and facilitate the simultaneous execution of multiple tasks,such as instance segmentation,scene classification and semantic segmentation.By leveraging these modules,distinct agents within multi-robot systems can effectively handle specialised tasks,thereby enhancing the overall system's flexibility and adaptability.This collaborative effort maximises the strengths of each robot,resulting in a more comprehensive understanding of environments.Extensive experiments on two public benchmark datasets demonstrate MPIL's competitive performance compared to state-of-the-art approaches,highlighting the effectiveness and robustness of our multi-robot system in complex indoor environments.展开更多
With the rapid urbanization and exponential population growth in China,two-wheeled vehicles have become a popular mode of transportation,particularly for short-distance travel.However,due to a lack of safety awareness...With the rapid urbanization and exponential population growth in China,two-wheeled vehicles have become a popular mode of transportation,particularly for short-distance travel.However,due to a lack of safety awareness,traffic violations by two-wheeled vehicle riders have become a widespread concern,contributing to urban traffic risks.Currently,significant human and material resources are being allocated to monitor and intercept non-compliant riders to ensure safe driving behavior.To enhance the safety,efficiency,and cost-effectiveness of traffic monitoring,automated detection systems based on image processing algorithms can be employed to identify traffic violations from eye-level video footage.In this study,we propose a robust detection algorithm specifically designed for two-wheeled vehicles,which serves as a fundamental step toward intelligent traffic monitoring.Our approach integrates a novel convolutional and attention mechanism to improve detection accuracy and efficiency.Additionally,we introduce a semi-supervised training strategy that leverages a large number of unlabeled images to enhance the model’s learning capability by extracting valuable background information.This method enables the model to generalize effectively to diverse urban environments and varying lighting conditions.We evaluate our proposed algorithm on a custom-built dataset,and experimental results demonstrate its superior performance,achieving an average precision(AP)of 95%and a recall(R)of 90.6%.Furthermore,the model maintains a computational efficiency of only 25.7 GFLOPs while achieving a high processing speed of 249 FPS,making it highly suitable for deployment on edge devices.Compared to existing detection methods,our approach significantly enhances the accuracy and robustness of two-wheeled vehicle identification while ensuring real-time performance.展开更多
Hepatitis is an infection that affects the liver through contaminated foods or blood transfusions,and it has many types,from normal to serious.Hepatitis is diagnosed through many blood tests and factors;Artificial Int...Hepatitis is an infection that affects the liver through contaminated foods or blood transfusions,and it has many types,from normal to serious.Hepatitis is diagnosed through many blood tests and factors;Artificial Intelligence(AI)techniques have played an important role in early diagnosis and help physicians make decisions.This study evaluated the performance of Machine Learning(ML)algorithms on the hepatitis data set.The dataset contains missing values that have been processed and outliers removed.The dataset was counterbalanced by the Synthetic Minority Over-sampling Technique(SMOTE).The features of the data set were processed in two ways:first,the application of the Recursive Feature Elimination(RFE)algorithm to arrange the percentage of contribution of each feature to the diagnosis of hepatitis,then selection of important features using the t-distributed Stochastic Neighbor Embedding(t-SNE)and Principal Component Analysis(PCA)algorithms.Second,the SelectKBest function was applied to give scores for each attribute,followed by the t-SNE and PCA algorithms.Finally,the classification algorithms K-Nearest Neighbors(KNN),Support Vector Machine(SVM),Artificial Neural Network(ANN),Decision Tree(DT),and Random Forest(RF)were fed by the dataset after processing the features in different methods are RFE with t-SNE and PCA and SelectKBest with t-SNE and PCA).All algorithms yielded promising results for diagnosing hepatitis data sets.The RF with RFE and PCA methods achieved accuracy,Precision,Recall,and AUC of 97.18%,96.72%,97.29%,and 94.2%,respectively,during the training phase.During the testing phase,it reached accuracy,Precision,Recall,and AUC by 96.31%,95.23%,97.11%,and 92.67%,respectively.展开更多
Nonlinear science is a fundamental area of physics research that investigates complex dynamical systems which are often characterized by high sensitivity and nonlinear behaviors.Numerical simulations play a pivotal ro...Nonlinear science is a fundamental area of physics research that investigates complex dynamical systems which are often characterized by high sensitivity and nonlinear behaviors.Numerical simulations play a pivotal role in nonlinear science,serving as a critical tool for revealing the underlying principles governing these systems.In addition,they play a crucial role in accelerating progress across various fields,such as climate modeling,weather forecasting,and fluid dynamics.However,their high computational cost limits their application in high-precision or long-duration simulations.In this study,we propose a novel data-driven approach for simulating complex physical systems,particularly turbulent phenomena.Specifically,we develop an efficient surrogate model based on the wavelet neural operator(WNO).Experimental results demonstrate that the enhanced WNO model can accurately simulate small-scale turbulent flows while using lower computational costs.In simulations of complex physical fields,the improved WNO model outperforms established deep learning models,such as U-Net,Res Net,and the Fourier neural operator(FNO),in terms of accuracy.Notably,the improved WNO model exhibits exceptional generalization capabilities,maintaining stable performance across a wide range of initial conditions and high-resolution scenarios without retraining.This study highlights the significant potential of the enhanced WNO model for simulating complex physical systems,providing strong evidence to support the development of more efficient,scalable,and high-precision simulation techniques.展开更多
Speech-face association aims to achieve identity matching between facial images and voice segments by aligning cross-modal features.Existing research primarily focuses on learning shared-space representations and comp...Speech-face association aims to achieve identity matching between facial images and voice segments by aligning cross-modal features.Existing research primarily focuses on learning shared-space representations and computing one-to-one similarities between cross-modal sample pairs to establish their correlation.However,these approaches do not fully account for intra-class variations between the modalities or the many-to-many relationships among cross-modal samples,which are crucial for robust association modeling.To address these challenges,we propose a novel framework that leverages global information to align voice and face embeddings while effectively correlating identity information embedded in both modalities.First,we jointly pre-train face recognition and speaker recognition networks to encode discriminative features from facial images and voice segments.This shared pre-training step ensures the extraction of complementary identity information across modalities.Subsequently,we introduce a cross-modal simplex center loss,which aligns samples with identity centers located at the vertices of a regular simplex inscribed on a hypersphere.This design enforces an equidistant and balanced distribution of identity embeddings,reducing intra-class variations.Furthermore,we employ an improved triplet center loss that emphasizes hard sample mining and optimizes inter-class separability,enhancing the model’s ability to generalize across challenging scenarios.Extensive experiments validate the effectiveness of our framework,demonstrating superior performance across various speech-face association tasks,including matching,verification,and retrieval.Notably,in the challenging gender-constrained matching task,our method achieves a remarkable accuracy of 79.22%,significantly outperforming existing approaches.These results highlight the potential of the proposed framework to advance the state of the art in cross-modal identity association.展开更多
Objective:This study aimed to develop and validate a predictive model for postoperative complications in gastrointestinal cancer patients using a large multicenter database,based on machine learning algorithms.Methods...Objective:This study aimed to develop and validate a predictive model for postoperative complications in gastrointestinal cancer patients using a large multicenter database,based on machine learning algorithms.Methods:We analyzed the clinicopathological data of 3,926 gastrointestinal cancer patients from the Prevalence of Abdominal Complications After GastroEnterological surgery(PACAGE)database,covering 20 medical centers from December 2018 to December 2020.The predictive performance was evaluated using receiver operating characteristic(ROC)curves and Brier Score.Results:The patients were divided into gastric(2,271 cases)and colorectal cancer(1,655 cases)groups and further divided into training and external validation sets.The overall postoperative complication rates for gastric and colorectal cancer groups were 18.1%and 14.8%,respectively.The most common complication was the intraabdominal infection in both gastric and colorectal cancer groups.In the training set,the Random Forest(RF)model predicted the highest mean area under the curve(AUC)values for overall complications and different types of complications,in both the gastric cancer group and the colorectal cancer group,with similar results obtained in the external validation set.ROC curve analysis showed good predictive performance of the RF model for overall and infectious complications.An application-based clinical tool was developed for easy application in clinical practice.Conclusions:This model demonstrated good predictive performance for overall and infectious complications based on the multi-center database,supporting clinical decision-making and personalized treatment strategies.展开更多
To capture the nonlinear dynamics and gain evolution in chirped pulse amplification(CPA)systems,the split-step Fourier method and the fourth-order Runge–Kutta method are integrated to iteratively address the generali...To capture the nonlinear dynamics and gain evolution in chirped pulse amplification(CPA)systems,the split-step Fourier method and the fourth-order Runge–Kutta method are integrated to iteratively address the generalized nonlinear Schrödinger equation and the rate equations.However,this approach is burdened by substantial computational demands,resulting in significant time expenditures.In the context of intelligent laser optimization and inverse design,the necessity for numerous simulations further exacerbates this issue,highlighting the need for fast and accurate simulation methodologies.Here,we introduce an end-to-end model augmented with active learning(E2E-AL)with decent generalization through different dedicated embedding methods over various parameters.On an identical computational platform,the artificial intelligence–driven model is 2000 times faster than the conventional simulation method.Benefiting from the active learning strategy,the E2E-AL model achieves decent precision with only two-thirds of the training samples compared with the case without such a strategy.Furthermore,we demonstrate a multi-objective inverse design of the CPA systems enabled by the E2E-AL model.The E2E-AL framework manifests the potential of becoming a standard approach for the rapid and accurate modeling of ultrafast lasers and is readily extended to simulate other complex systems.展开更多
The manufacturing processes of casing rings are prone to multi-type defects such as holes,cracks,and porosity,so ultrasonic testing is vital for the quality of aeroengine.Conventional ultrasonic testing requires manua...The manufacturing processes of casing rings are prone to multi-type defects such as holes,cracks,and porosity,so ultrasonic testing is vital for the quality of aeroengine.Conventional ultrasonic testing requires manual analysis,which is susceptible to human omission,inconsistent results,and time-consumption.In this paper,a method for automated detection of defects is proposed for the ultrasonic Total Focusing Method(TFM)inspection of casing rings based on deep learning.First,the original datasets of defect images are established,and the Mask R-CNN is used to increase the number of defects in a single image.Then,the YOLOX-S-improved lightweight model is proposed,and the feature extraction network is replaced by Faster Net to reduce redundant computations.The Super-Resolution Generative Adversarial Network(SRGAN)and Convolutional Block Attention Module(CBAM)are integrated to improve the identification precision.Finally,a new test dataset is created by ultrasonic TFM inspection of an aeroengine casing ring.The results show that the mean of Average Precision(m AP)of the YOLOX-S-improved model reaches 99.17%,and the corresponding speed reaches 77.6 FPS.This study indicates that the YOLOX-S-improved model performs better than conventional object detection models.And the generalization ability of the proposed model is verified by ultrasonic B-scan images.展开更多
Landslide susceptibility evaluation plays an important role in disaster prevention and reduction.Feature-based transfer learning(TL)is an effective method for solving landslide susceptibility mapping(LSM)in target reg...Landslide susceptibility evaluation plays an important role in disaster prevention and reduction.Feature-based transfer learning(TL)is an effective method for solving landslide susceptibility mapping(LSM)in target regions with no available samples.However,as the study area expands,the distribution of land-slide types and triggering mechanisms becomes more diverse,leading to performance degradation in models relying on landslide evaluation knowledge from a single source domain due to domain feature shift.To address this,this study proposes a Multi-source Domain Adaptation Convolutional Neural Network(MDACNN),which combines the landslide prediction knowledge learned from two source domains to perform cross-regional LSM in complex large-scale areas.The method is validated through case studies in three regions located in southeastern coastal China and compared with single-source domain TL models(TCA-based models).The results demonstrate that MDACNN effectively integrates transfer knowledge from multiple source domains to learn diverse landslide-triggering mechanisms,thereby significantly reducing prediction bias inherent to single-source domain TL models,achieving an average improvement of 16.58%across all metrics.Moreover,the landslide susceptibility maps gener-ated by MDACNN accurately quantify the spatial distribution of landslide risks in the target area,provid-ing a powerful scientific and technological tool for landslide disaster management and prevention.展开更多
BACKGROUND Neoadjuvant therapy prior to surgery plays a critical role in improving the prognosis of patients with unresectable or locally advanced gastric cancer(GC).Postoperative complications,particularly those clas...BACKGROUND Neoadjuvant therapy prior to surgery plays a critical role in improving the prognosis of patients with unresectable or locally advanced gastric cancer(GC).Postoperative complications,particularly those classified as Clavien-Dindo grade≥II,remain a major concern for surgeons.In recent years machine learning(ML)has emerged as a prominent approach for disease diagnosis and prediction.However,studies on both postoperative complications and ML in patients with GC receiving neoadjuvant therapy remain limited.AIM To develop an ML model to predict Clavien-Dindo grade≥II complications in patients with GC after neoadjuvant therapy and laparoscopic gastrectomy.METHODS Clinical data were collected from 455 patients with GC who underwent neoadjuvant therapy followed by laparoscopic gastrectomy at Peking Union Medical College Hospital(2014-2024).Potential predictors were identified through univariate analysis and least absolute shrinkage and selection operator regression.Six ML algorithms including XGBoost,random forest,neural network ensemble(NNE),logistic regression,GLMnet,and decision tree were trained and optimized using nested cross-validation.Model performance was evaluated using the area under the receiver operating characteristic curve,decision curve analysis,and calibration curves.RESULTS A total of 455 patients were included of whom 69(15.16%)developed Clavien-Dindo grade≥II complications.The predictive model was constructed using seven variables,including smoking status,Nutritional Risk Screening-2002 score,American Society of Anesthesiologists classification,neoadjuvant therapy,surgical approach,operating time,and intraoperative blood loss.Among the six models the NNE model outperformed the others,achieving the highest area under the receiver operating characteristic curve(0.789,0.739-0.840)and demonstrating superior discrimination,clinical utility,and calibration.CONCLUSION The NNE-based prediction model effectively identified patients with GC at high risk of Clavien-Dindo grade≥II complications after neoadjuvant therapy and laparoscopic gastrectomy.展开更多
Porters play a crucial role in hospitals because they ensure the efficient transportation of patients,medical equipment,and vital documents.Despite its importance,there is a lack of research addressing the prediction ...Porters play a crucial role in hospitals because they ensure the efficient transportation of patients,medical equipment,and vital documents.Despite its importance,there is a lack of research addressing the prediction of completion times for porter tasks.To address this gap,we utilized real-world porter delivery data from Taiwan University Hospital,China,Yunlin Branch,Taiwan Region of China.We first identified key features that can influence the duration of porter tasks.We then employed three widely-used machine learning algorithms:decision tree,random forest,and gradient boosting.To leverage the strengths of each algorithm,we finally adopted an ensemble modeling approach that aggregates their individual predictions.Our experimental results show that the proposed ensemble model can achieve a mean absolute error of 3 min in predicting task response time and 4.42 min in task completion time.The prediction error is around 50%lower compared to using only the historical average.These results demonstrate that our method significantly improves the accuracy of porter task time prediction,supporting better resource planning and patient care.It helps ward staff streamline workflows by reducing delays,enables porter managers to allocate resources more effectively,and shortens patient waiting times,contributing to a better care experience.展开更多
The deep learning algorithm,which has been increasingly applied in the field of petroleum geophysical prospecting,has achieved good results in improving efficiency and accuracy based on test applications.To play a gre...The deep learning algorithm,which has been increasingly applied in the field of petroleum geophysical prospecting,has achieved good results in improving efficiency and accuracy based on test applications.To play a greater role in actual production,these algorithm modules must be integrated into software systems and used more often in actual production projects.Deep learning frameworks,such as TensorFlow and PyTorch,basically take Python as the core architecture,while the application program mainly uses Java,C#,and other programming languages.During integration,the seismic data read by the Java and C#data interfaces must be transferred to the Python main program module.The data exchange methods between Java,C#,and Python include shared memory,shared directory,and so on.However,these methods have the disadvantages of low transmission efficiency and unsuitability for asynchronous networks.Considering the large volume of seismic data and the need for network support for deep learning,this paper proposes a method of transmitting seismic data based on Socket.By maximizing Socket’s cross-network and efficient longdistance transmission,this approach solves the problem of inefficient transmission of underlying data while integrating the deep learning algorithm module into a software system.Furthermore,the actual production application shows that this method effectively solves the shortage of data transmission in shared memory,shared directory,and other modes while simultaneously improving the transmission efficiency of massive seismic data across modules at the bottom of the software.展开更多
Neutron time-of-flight(ToF)measurement is a highly accurate method for obtaining the kinetic energy of a neutron by measuring its velocity,but requires precise acquisition of the neutron signal arrival time.However,th...Neutron time-of-flight(ToF)measurement is a highly accurate method for obtaining the kinetic energy of a neutron by measuring its velocity,but requires precise acquisition of the neutron signal arrival time.However,the high hardware costs and data burden associated with the acquisition of neutron ToF signals pose significant challenges.Higher sampling rates increase the data volume,data processing,and storage hardware costs.Compressed sampling can address these challenges,but it faces issues regarding optimal sampling efficiency and high-quality reconstructed signals.This paper proposes a revolutionary deep learning-based compressed sampling(DL-CS)algorithm for reconstructing neutron ToF signals that outperform traditional compressed sampling methods.This approach comprises four modules:random projection,rising dimensions,initial reconstruction,and final reconstruction.Initially,the technique adaptively compresses neutron ToF signals sequentially using three convolutional layers,replacing random measurement matrices in traditional compressed sampling theory.Subsequently,the signals are reconstructed using a modified inception module,long short-term memory,and self-attention.The performance of this deep compressed sampling method was quantified using the percentage root-mean-square difference,correlation coefficient,and reconstruction time.Experimental results showed that our proposed DL-CS approach can significantly enhance signal quality compared with other compressed sampling methods.This is evidenced by a percentage root-mean-square difference,correlation coefficient,and reconstruction time results of 5%,0.9988,and 0.0108 s,respectively,obtained for sampling rates below 10%for the neutron ToF signal generated using an electron-beam-driven photoneutron source.The results showed that the proposed DL-CS approach significantly improves the signal quality compared with other compressed sampling methods,exhibiting excellent reconstruction accuracy and speed.展开更多
Cooperative task assignment is one of the key research focuses in the field of unmanned aerial vehicles(UAVs). In this paper, an energy learning hyper-heuristic(EL-HH) algorithm is proposed to address the cooperative ...Cooperative task assignment is one of the key research focuses in the field of unmanned aerial vehicles(UAVs). In this paper, an energy learning hyper-heuristic(EL-HH) algorithm is proposed to address the cooperative task assignment problem of heterogeneous UAVs under complex constraints. First, a mathematical model is designed to define the scenario, complex constraints, and objective function of the problem. Then, the scheme encoding, the EL-HH strategy, multiple optimization operators, and the task sequence and time adjustment strategies are designed in the EL-HH algorithm. The scheme encoding is designed with three layers: task sequence, UAV sequence, and waiting time. The EL-HH strategy applies an energy learning method to adaptively adjust the energies of operators, thereby facilitating the selection and application of operators. Multiple optimization operators can update schemes in different ways, enabling the algorithm to fully explore the solution space. Afterward, the task order and time adjustment strategies are designed to adjust task order and insert waiting time. Through the iterative optimization process, a satisfactory assignment scheme is ultimately produced. Finally, simulation and experiment verify the effectiveness of the proposed algorithm.展开更多
The electromagnetic pulse valve,as a key component in baghouse dust removal systems,plays a crucial role in the performance of the system.However,despite the promising results of intelligent fault diagnosis methods ba...The electromagnetic pulse valve,as a key component in baghouse dust removal systems,plays a crucial role in the performance of the system.However,despite the promising results of intelligent fault diagnosis methods based on extensive data in diagnosing electromagnetic valves,real-world diagnostic scenarios still face numerous challenges.Collecting fault data for electromagnetic pulse valves is not only time-consuming but also costly,making it difficult to obtain sufficient fault data in advance,which poses challenges for small sample fault diagnosis.To address this issue,this paper proposes a fault diagnosis method for electromagnetic pulse valves based on deep transfer learning and simulated data.This method achieves effective transfer from simulated data to real data through four parameter transfer strategies,which combine parameter freezing and fine-tuning operations.Furthermore,this paper identifies a parameter transfer strategy that simultaneously fine-tunes the feature extractor and classifier,and introduces an attention mechanism to integrate fault features,thereby enhancing the correlation and information complementarity among multi-sensor data.The effectiveness of the proposed method is evaluated through two fault diagnosis cases under different operating conditions.In this study,small sample data accounted for 7.9%and 8.2%of the total dataset,and the experimental results showed transfer accuracies of 93.5%and 94.2%,respectively,validating the reliability and effectiveness of the method under small sample conditions.展开更多
基金supported by National Natural Science Foundation of China(62466045)Inner Mongolia Natural Science Foundation Project(2021LHMS06003)Inner Mongolia University Basic Research Business Fee Project(114).
文摘Graph Federated Learning(GFL)has shown great potential in privacy protection and distributed intelligence through distributed collaborative training of graph-structured data without sharing raw information.However,existing GFL approaches often lack the capability for comprehensive feature extraction and adaptive optimization,particularly in non-independent and identically distributed(NON-IID)scenarios where balancing global structural understanding and local node-level detail remains a challenge.To this end,this paper proposes a novel framework called GFL-SAR(Graph Federated Collaborative Learning Framework Based on Structural Amplification and Attention Refinement),which enhances the representation learning capability of graph data through a dual-branch collaborative design.Specifically,we propose the Structural Insight Amplifier(SIA),which utilizes an improved Graph Convolutional Network(GCN)to strengthen structural awareness and improve modeling of topological patterns.In parallel,we propose the Attentive Relational Refiner(ARR),which employs an enhanced Graph Attention Network(GAT)to perform fine-grained modeling of node relationships and neighborhood features,thereby improving the expressiveness of local interactions and preserving critical contextual information.GFL-SAR effectively integrates multi-scale features from every branch via feature fusion and federated optimization,thereby addressing existing GFL limitations in structural modeling and feature representation.Experiments on standard benchmark datasets including Cora,Citeseer,Polblogs,and Cora_ML demonstrate that GFL-SAR achieves superior performance in classification accuracy,convergence speed,and robustness compared to existing methods,confirming its effectiveness and generalizability in GFL tasks.
基金supported by the National Natural Science Foundation of China(No.22276139)the Shanghai’s Municipal State-owned Assets Supervision and Administration Commission(No.2022028).
文摘To better understand the migration behavior of plastic fragments in the environment,development of rapid non-destructive methods for in-situ identification and characterization of plastic fragments is necessary.However,most of the studies had focused only on colored plastic fragments,ignoring colorless plastic fragments and the effects of different environmental media(backgrounds),thus underestimating their abundance.To address this issue,the present study used near-infrared spectroscopy to compare the identification of colored and colorless plastic fragments based on partial least squares-discriminant analysis(PLS-DA),extreme gradient boost,support vector machine and random forest classifier.The effects of polymer color,type,thickness,and background on the plastic fragments classification were evaluated.PLS-DA presented the best and most stable outcome,with higher robustness and lower misclassification rate.All models frequently misinterpreted colorless plastic fragments and its background when the fragment thickness was less than 0.1mm.A two-stage modeling method,which first distinguishes the plastic types and then identifies colorless plastic fragments that had been misclassified as background,was proposed.The method presented an accuracy higher than 99%in different backgrounds.In summary,this study developed a novel method for rapid and synchronous identification of colored and colorless plastic fragments under complex environmental backgrounds.
基金funded by the China Chongqing Municipal Science and Technology Bureau,grant numbers CSTB2024TIAD-CYKJCXX0009,CSTB2024NSCQ-LZX0043,CSTB2022NSCQ-MSX0288Chongqing Municipal Commission of Housing and Urban-Rural Development,grant number CKZ2024-87+3 种基金the Chongqing University of Technology Graduate Education High-Quality Development Project,grant number gzlsz202401the Chongqing University of Technology—Chongqing LINGLUE Technology Co.,Ltd.Electronic Information(Artificial Intelligence)Graduate Joint Training Basethe Postgraduate Education and Teaching Reform Research Project in Chongqing,grant number yjg213116the Chongqing University of Technology-CISDI Chongqing Information Technology Co.,Ltd.Computer Technology Graduate Joint Training Base.
文摘Although conventional object detection methods achieve high accuracy through extensively annotated datasets,acquiring such large-scale labeled data remains challenging and cost-prohibitive in numerous real-world applications.Few-shot object detection presents a new research idea that aims to localize and classify objects in images using only limited annotated examples.However,the inherent challenge in few-shot object detection lies in the insufficient sample diversity to fully characterize the sample feature distribution,which consequently impacts model performance.Inspired by contrastive learning principles,we propose an Implicit Feature Contrastive Learning(IFCL)module to address this limitation and augment feature diversity for more robust representational learning.This module generates augmented support sample features in a mixed feature space and implicitly contrasts them with query Region of Interest(RoI)features.This approach facilitates more comprehensive learning of both intra-class feature similarity and inter-class feature diversity,thereby enhancing the model’s object classification and localization capabilities.Extensive experiments on PASCAL VOC show that our method achieves a respective improvement of 3.2%,1.8%,and 2.3%on 10-shot of three Novel Sets compared to the baseline model FPD.
基金funded by Hanshan Normal University School-Level Research Initiation Program(grant numbers QD202244QD2024207)+3 种基金the Guangdong Higher Education Society’s“Fourteenth Five-Year”Plan 2024 Higher Education Research(grant number 24GYB43)the 2024 Guangdong Provincial Undergraduate Teaching Quality and Teaching Reform Engineering Project:Excellence Program for Cultivating Publicly-Funded Pre-service Teachers for Primary Education in the Context of Rural Revitalizationthe Hanshan Normal University Guangdong East Regional Education Collaborative Innovation Research Centerfunded by these sources.
文摘This study utilized a sequential mediating model to examine the role of motivation to learn and transfer selfefficacy in the relationships between perceived content validity,mentoring function,continuous learning work culture and intention to transfer learning.The sample comprized 429 final-year apprentices in Guangdong province,China(females=69.9%,Engineering&Medicine=69%,mean age=20.99,SD=1.60).The apprentices completed standardized measures of motivation to learn,transfer self-efficacy perceived content validity,mentoring function,and continuous learning work culture.Structural equation modeling was used to analyze the data.Results showed perceived content validity,mentoring function,continuous learning culture to predict intention to transfer learning.Of these factors,perceived content validity was the strongest predictor of intention to transfer learning.Of these factors,perceived content validity was the most influential predictor of intention to transfer learning.The motivation to learn and transfer self-efficacy sequentially mediated the relationship between mentoring function and intention to learning transfer to be stronger than by either alone.Although perceived content validity and continuous learning culture exhibited no significant direct effects on intention to transfer learning,they demonstrated positive indirect associations with intention to transfer via motivation to learn and transfer self-efficacy.These study findings extend the applications of the learning transfer framework to individuals undergoing apprenticeship training which also would apply to other a long-term work-based learning programs.
文摘Hepatocellular carcinoma(HCC)remains a leading cause of cancer-related mortality globally,necessitating advanced diagnostic tools to improve early detection and personalized targeted therapy.This review synthesizes evidence on explainable ensemble learning approaches for HCC classification,emphasizing their integration with clinical workflows and multi-omics data.A systematic analysis[including datasets such as The Cancer Genome Atlas,Gene Expression Omnibus,and the Surveillance,Epidemiology,and End Results(SEER)datasets]revealed that explainable ensemble learning models achieve high diagnostic accuracy by combining clinical features,serum biomarkers such as alpha-fetoprotein,imaging features such as computed tomography and magnetic resonance imaging,and genomic data.For instance,SHapley Additive exPlanations(SHAP)-based random forests trained on NCBI GSE14520 microarray data(n=445)achieved 96.53%accuracy,while stacking ensembles applied to the SEER program data(n=1897)demonstrated an area under the receiver operating characteristic curve of 0.779 for mortality prediction.Despite promising results,challenges persist,including the computational costs of SHAP and local interpretable model-agnostic explanations analyses(e.g.,TreeSHAP requiring distributed computing for metabolomics datasets)and dataset biases(e.g.,SEER’s Western population dominance limiting generalizability).Future research must address inter-cohort heterogeneity,standardize explainability metrics,and prioritize lightweight surrogate models for resource-limited settings.This review presents the potential of explainable ensemble learning frameworks to bridge the gap between predictive accuracy and clinical interpretability,though rigorous validation in independent,multi-center cohorts is critical for real-world deployment.
基金supported by the National Natural Science Foundation of China under Grant 62373009.
文摘Indoor scene semantic segmentation is essential for enabling robots to understand and interact with their environments effectively.However,numerous challenges remain unresolved,particularly in single-robot systems,which often struggle with the complexity and variability of indoor scenes.To address these limitations,we introduce a novel multi-robot collaborative framework based on multiplex interactive learning(MPIL)in which each robot specialises in a distinct visual task within a unified multitask architecture.During training,the framework employs task-specific decoders and cross-task feature sharing to enhance collaborative optimisation.At inference time,robots operate independently with optimised models,enabling scalable,asynchronous and efficient deployment in real-world scenarios.Specifically,MPIL employs specially designed modules that integrate RGB and depth data,refine feature representations and facilitate the simultaneous execution of multiple tasks,such as instance segmentation,scene classification and semantic segmentation.By leveraging these modules,distinct agents within multi-robot systems can effectively handle specialised tasks,thereby enhancing the overall system's flexibility and adaptability.This collaborative effort maximises the strengths of each robot,resulting in a more comprehensive understanding of environments.Extensive experiments on two public benchmark datasets demonstrate MPIL's competitive performance compared to state-of-the-art approaches,highlighting the effectiveness and robustness of our multi-robot system in complex indoor environments.
基金supported by the Natural Science Foundation Project of Fujian Province,China(Grant No.2023J011439 and No.2019J01859).
文摘With the rapid urbanization and exponential population growth in China,two-wheeled vehicles have become a popular mode of transportation,particularly for short-distance travel.However,due to a lack of safety awareness,traffic violations by two-wheeled vehicle riders have become a widespread concern,contributing to urban traffic risks.Currently,significant human and material resources are being allocated to monitor and intercept non-compliant riders to ensure safe driving behavior.To enhance the safety,efficiency,and cost-effectiveness of traffic monitoring,automated detection systems based on image processing algorithms can be employed to identify traffic violations from eye-level video footage.In this study,we propose a robust detection algorithm specifically designed for two-wheeled vehicles,which serves as a fundamental step toward intelligent traffic monitoring.Our approach integrates a novel convolutional and attention mechanism to improve detection accuracy and efficiency.Additionally,we introduce a semi-supervised training strategy that leverages a large number of unlabeled images to enhance the model’s learning capability by extracting valuable background information.This method enables the model to generalize effectively to diverse urban environments and varying lighting conditions.We evaluate our proposed algorithm on a custom-built dataset,and experimental results demonstrate its superior performance,achieving an average precision(AP)of 95%and a recall(R)of 90.6%.Furthermore,the model maintains a computational efficiency of only 25.7 GFLOPs while achieving a high processing speed of 249 FPS,making it highly suitable for deployment on edge devices.Compared to existing detection methods,our approach significantly enhances the accuracy and robustness of two-wheeled vehicle identification while ensuring real-time performance.
基金funded by Scientific Research Deanship at University of Ha’il,Saudi Arabia,through project number GR-24009.
文摘Hepatitis is an infection that affects the liver through contaminated foods or blood transfusions,and it has many types,from normal to serious.Hepatitis is diagnosed through many blood tests and factors;Artificial Intelligence(AI)techniques have played an important role in early diagnosis and help physicians make decisions.This study evaluated the performance of Machine Learning(ML)algorithms on the hepatitis data set.The dataset contains missing values that have been processed and outliers removed.The dataset was counterbalanced by the Synthetic Minority Over-sampling Technique(SMOTE).The features of the data set were processed in two ways:first,the application of the Recursive Feature Elimination(RFE)algorithm to arrange the percentage of contribution of each feature to the diagnosis of hepatitis,then selection of important features using the t-distributed Stochastic Neighbor Embedding(t-SNE)and Principal Component Analysis(PCA)algorithms.Second,the SelectKBest function was applied to give scores for each attribute,followed by the t-SNE and PCA algorithms.Finally,the classification algorithms K-Nearest Neighbors(KNN),Support Vector Machine(SVM),Artificial Neural Network(ANN),Decision Tree(DT),and Random Forest(RF)were fed by the dataset after processing the features in different methods are RFE with t-SNE and PCA and SelectKBest with t-SNE and PCA).All algorithms yielded promising results for diagnosing hepatitis data sets.The RF with RFE and PCA methods achieved accuracy,Precision,Recall,and AUC of 97.18%,96.72%,97.29%,and 94.2%,respectively,during the training phase.During the testing phase,it reached accuracy,Precision,Recall,and AUC by 96.31%,95.23%,97.11%,and 92.67%,respectively.
基金Project supported by the National Natural Science Foundation of China(Grant Nos.42005003 and 41475094)。
文摘Nonlinear science is a fundamental area of physics research that investigates complex dynamical systems which are often characterized by high sensitivity and nonlinear behaviors.Numerical simulations play a pivotal role in nonlinear science,serving as a critical tool for revealing the underlying principles governing these systems.In addition,they play a crucial role in accelerating progress across various fields,such as climate modeling,weather forecasting,and fluid dynamics.However,their high computational cost limits their application in high-precision or long-duration simulations.In this study,we propose a novel data-driven approach for simulating complex physical systems,particularly turbulent phenomena.Specifically,we develop an efficient surrogate model based on the wavelet neural operator(WNO).Experimental results demonstrate that the enhanced WNO model can accurately simulate small-scale turbulent flows while using lower computational costs.In simulations of complex physical fields,the improved WNO model outperforms established deep learning models,such as U-Net,Res Net,and the Fourier neural operator(FNO),in terms of accuracy.Notably,the improved WNO model exhibits exceptional generalization capabilities,maintaining stable performance across a wide range of initial conditions and high-resolution scenarios without retraining.This study highlights the significant potential of the enhanced WNO model for simulating complex physical systems,providing strong evidence to support the development of more efficient,scalable,and high-precision simulation techniques.
基金funded by the Scientific Funding for China Academy of Railway Sciences Corporation Limited,China(No.2023YJ125).
文摘Speech-face association aims to achieve identity matching between facial images and voice segments by aligning cross-modal features.Existing research primarily focuses on learning shared-space representations and computing one-to-one similarities between cross-modal sample pairs to establish their correlation.However,these approaches do not fully account for intra-class variations between the modalities or the many-to-many relationships among cross-modal samples,which are crucial for robust association modeling.To address these challenges,we propose a novel framework that leverages global information to align voice and face embeddings while effectively correlating identity information embedded in both modalities.First,we jointly pre-train face recognition and speaker recognition networks to encode discriminative features from facial images and voice segments.This shared pre-training step ensures the extraction of complementary identity information across modalities.Subsequently,we introduce a cross-modal simplex center loss,which aligns samples with identity centers located at the vertices of a regular simplex inscribed on a hypersphere.This design enforces an equidistant and balanced distribution of identity embeddings,reducing intra-class variations.Furthermore,we employ an improved triplet center loss that emphasizes hard sample mining and optimizes inter-class separability,enhancing the model’s ability to generalize across challenging scenarios.Extensive experiments validate the effectiveness of our framework,demonstrating superior performance across various speech-face association tasks,including matching,verification,and retrieval.Notably,in the challenging gender-constrained matching task,our method achieves a remarkable accuracy of 79.22%,significantly outperforming existing approaches.These results highlight the potential of the proposed framework to advance the state of the art in cross-modal identity association.
基金supported by the Natural Science Foundation of Fujian Province(No.2022J01755)。
文摘Objective:This study aimed to develop and validate a predictive model for postoperative complications in gastrointestinal cancer patients using a large multicenter database,based on machine learning algorithms.Methods:We analyzed the clinicopathological data of 3,926 gastrointestinal cancer patients from the Prevalence of Abdominal Complications After GastroEnterological surgery(PACAGE)database,covering 20 medical centers from December 2018 to December 2020.The predictive performance was evaluated using receiver operating characteristic(ROC)curves and Brier Score.Results:The patients were divided into gastric(2,271 cases)and colorectal cancer(1,655 cases)groups and further divided into training and external validation sets.The overall postoperative complication rates for gastric and colorectal cancer groups were 18.1%and 14.8%,respectively.The most common complication was the intraabdominal infection in both gastric and colorectal cancer groups.In the training set,the Random Forest(RF)model predicted the highest mean area under the curve(AUC)values for overall complications and different types of complications,in both the gastric cancer group and the colorectal cancer group,with similar results obtained in the external validation set.ROC curve analysis showed good predictive performance of the RF model for overall and infectious complications.An application-based clinical tool was developed for easy application in clinical practice.Conclusions:This model demonstrated good predictive performance for overall and infectious complications based on the multi-center database,supporting clinical decision-making and personalized treatment strategies.
基金supported by the National Natural Science Foundation of China(Grant Nos.62227821,62025503,and 62205199).
文摘To capture the nonlinear dynamics and gain evolution in chirped pulse amplification(CPA)systems,the split-step Fourier method and the fourth-order Runge–Kutta method are integrated to iteratively address the generalized nonlinear Schrödinger equation and the rate equations.However,this approach is burdened by substantial computational demands,resulting in significant time expenditures.In the context of intelligent laser optimization and inverse design,the necessity for numerous simulations further exacerbates this issue,highlighting the need for fast and accurate simulation methodologies.Here,we introduce an end-to-end model augmented with active learning(E2E-AL)with decent generalization through different dedicated embedding methods over various parameters.On an identical computational platform,the artificial intelligence–driven model is 2000 times faster than the conventional simulation method.Benefiting from the active learning strategy,the E2E-AL model achieves decent precision with only two-thirds of the training samples compared with the case without such a strategy.Furthermore,we demonstrate a multi-objective inverse design of the CPA systems enabled by the E2E-AL model.The E2E-AL framework manifests the potential of becoming a standard approach for the rapid and accurate modeling of ultrafast lasers and is readily extended to simulate other complex systems.
基金supported by the Postdoctoral Fellowship Program of CPSF,China(No.GZC20232015)the China Postdoctoral Science Foundation(No.2024M752499)+3 种基金the Postdoctoral Project of Hubei Province,China(No.2024HBBHCXA076)the Wuhan East Lake New Technology Development Zone Open List Project,China(No.2022KJB128)the National Natural Science Foundation of China(No.51875428)the Fundamental Research Funds for the Central Universities,China(No.104972024RSCbs0013)。
文摘The manufacturing processes of casing rings are prone to multi-type defects such as holes,cracks,and porosity,so ultrasonic testing is vital for the quality of aeroengine.Conventional ultrasonic testing requires manual analysis,which is susceptible to human omission,inconsistent results,and time-consumption.In this paper,a method for automated detection of defects is proposed for the ultrasonic Total Focusing Method(TFM)inspection of casing rings based on deep learning.First,the original datasets of defect images are established,and the Mask R-CNN is used to increase the number of defects in a single image.Then,the YOLOX-S-improved lightweight model is proposed,and the feature extraction network is replaced by Faster Net to reduce redundant computations.The Super-Resolution Generative Adversarial Network(SRGAN)and Convolutional Block Attention Module(CBAM)are integrated to improve the identification precision.Finally,a new test dataset is created by ultrasonic TFM inspection of an aeroengine casing ring.The results show that the mean of Average Precision(m AP)of the YOLOX-S-improved model reaches 99.17%,and the corresponding speed reaches 77.6 FPS.This study indicates that the YOLOX-S-improved model performs better than conventional object detection models.And the generalization ability of the proposed model is verified by ultrasonic B-scan images.
基金the National Natural Science Foundation of China(Grant No.42301002,and 52109118)Fujian Provincial Water Resources Science and Technology Project(Grant No.MSK202524)Guidance fund for Science and Technology Program,Fujian province(Grant No.2024Y0002).
文摘Landslide susceptibility evaluation plays an important role in disaster prevention and reduction.Feature-based transfer learning(TL)is an effective method for solving landslide susceptibility mapping(LSM)in target regions with no available samples.However,as the study area expands,the distribution of land-slide types and triggering mechanisms becomes more diverse,leading to performance degradation in models relying on landslide evaluation knowledge from a single source domain due to domain feature shift.To address this,this study proposes a Multi-source Domain Adaptation Convolutional Neural Network(MDACNN),which combines the landslide prediction knowledge learned from two source domains to perform cross-regional LSM in complex large-scale areas.The method is validated through case studies in three regions located in southeastern coastal China and compared with single-source domain TL models(TCA-based models).The results demonstrate that MDACNN effectively integrates transfer knowledge from multiple source domains to learn diverse landslide-triggering mechanisms,thereby significantly reducing prediction bias inherent to single-source domain TL models,achieving an average improvement of 16.58%across all metrics.Moreover,the landslide susceptibility maps gener-ated by MDACNN accurately quantify the spatial distribution of landslide risks in the target area,provid-ing a powerful scientific and technological tool for landslide disaster management and prevention.
基金Supported by the National Key Research and Development Program of China,No.2022YFF1100404the National High Level Hospital Clinical Research Funding of China,No.2022-PUMCH-B-005.
文摘BACKGROUND Neoadjuvant therapy prior to surgery plays a critical role in improving the prognosis of patients with unresectable or locally advanced gastric cancer(GC).Postoperative complications,particularly those classified as Clavien-Dindo grade≥II,remain a major concern for surgeons.In recent years machine learning(ML)has emerged as a prominent approach for disease diagnosis and prediction.However,studies on both postoperative complications and ML in patients with GC receiving neoadjuvant therapy remain limited.AIM To develop an ML model to predict Clavien-Dindo grade≥II complications in patients with GC after neoadjuvant therapy and laparoscopic gastrectomy.METHODS Clinical data were collected from 455 patients with GC who underwent neoadjuvant therapy followed by laparoscopic gastrectomy at Peking Union Medical College Hospital(2014-2024).Potential predictors were identified through univariate analysis and least absolute shrinkage and selection operator regression.Six ML algorithms including XGBoost,random forest,neural network ensemble(NNE),logistic regression,GLMnet,and decision tree were trained and optimized using nested cross-validation.Model performance was evaluated using the area under the receiver operating characteristic curve,decision curve analysis,and calibration curves.RESULTS A total of 455 patients were included of whom 69(15.16%)developed Clavien-Dindo grade≥II complications.The predictive model was constructed using seven variables,including smoking status,Nutritional Risk Screening-2002 score,American Society of Anesthesiologists classification,neoadjuvant therapy,surgical approach,operating time,and intraoperative blood loss.Among the six models the NNE model outperformed the others,achieving the highest area under the receiver operating characteristic curve(0.789,0.739-0.840)and demonstrating superior discrimination,clinical utility,and calibration.CONCLUSION The NNE-based prediction model effectively identified patients with GC at high risk of Clavien-Dindo grade≥II complications after neoadjuvant therapy and laparoscopic gastrectomy.
基金supported by National Taiwan University Hospital Yunlin Branch Project NTUHYL 110.C018National Science and Technology Council,Taiwan.
文摘Porters play a crucial role in hospitals because they ensure the efficient transportation of patients,medical equipment,and vital documents.Despite its importance,there is a lack of research addressing the prediction of completion times for porter tasks.To address this gap,we utilized real-world porter delivery data from Taiwan University Hospital,China,Yunlin Branch,Taiwan Region of China.We first identified key features that can influence the duration of porter tasks.We then employed three widely-used machine learning algorithms:decision tree,random forest,and gradient boosting.To leverage the strengths of each algorithm,we finally adopted an ensemble modeling approach that aggregates their individual predictions.Our experimental results show that the proposed ensemble model can achieve a mean absolute error of 3 min in predicting task response time and 4.42 min in task completion time.The prediction error is around 50%lower compared to using only the historical average.These results demonstrate that our method significantly improves the accuracy of porter task time prediction,supporting better resource planning and patient care.It helps ward staff streamline workflows by reducing delays,enables porter managers to allocate resources more effectively,and shortens patient waiting times,contributing to a better care experience.
基金supported by the PetroChina Prospective,Basic,and Strategic Technology Research Project(No.2021ZG03-02 and No.2023DJ8402)。
文摘The deep learning algorithm,which has been increasingly applied in the field of petroleum geophysical prospecting,has achieved good results in improving efficiency and accuracy based on test applications.To play a greater role in actual production,these algorithm modules must be integrated into software systems and used more often in actual production projects.Deep learning frameworks,such as TensorFlow and PyTorch,basically take Python as the core architecture,while the application program mainly uses Java,C#,and other programming languages.During integration,the seismic data read by the Java and C#data interfaces must be transferred to the Python main program module.The data exchange methods between Java,C#,and Python include shared memory,shared directory,and so on.However,these methods have the disadvantages of low transmission efficiency and unsuitability for asynchronous networks.Considering the large volume of seismic data and the need for network support for deep learning,this paper proposes a method of transmitting seismic data based on Socket.By maximizing Socket’s cross-network and efficient longdistance transmission,this approach solves the problem of inefficient transmission of underlying data while integrating the deep learning algorithm module into a software system.Furthermore,the actual production application shows that this method effectively solves the shortage of data transmission in shared memory,shared directory,and other modes while simultaneously improving the transmission efficiency of massive seismic data across modules at the bottom of the software.
基金supported by the National Defense Technology Foundation Program of China(No.JSJT2022209A001-3)Sichuan Science and Technology Program(No.2021JDRC0011)+1 种基金Nuclear Energy Development Research Program of China(Research on High Energy X-ray Imaging of Nuclear Fuel)Scientific Research and Innovation Team Program of Sichuan University of Science and Engineering(No.SUSE652A001).
文摘Neutron time-of-flight(ToF)measurement is a highly accurate method for obtaining the kinetic energy of a neutron by measuring its velocity,but requires precise acquisition of the neutron signal arrival time.However,the high hardware costs and data burden associated with the acquisition of neutron ToF signals pose significant challenges.Higher sampling rates increase the data volume,data processing,and storage hardware costs.Compressed sampling can address these challenges,but it faces issues regarding optimal sampling efficiency and high-quality reconstructed signals.This paper proposes a revolutionary deep learning-based compressed sampling(DL-CS)algorithm for reconstructing neutron ToF signals that outperform traditional compressed sampling methods.This approach comprises four modules:random projection,rising dimensions,initial reconstruction,and final reconstruction.Initially,the technique adaptively compresses neutron ToF signals sequentially using three convolutional layers,replacing random measurement matrices in traditional compressed sampling theory.Subsequently,the signals are reconstructed using a modified inception module,long short-term memory,and self-attention.The performance of this deep compressed sampling method was quantified using the percentage root-mean-square difference,correlation coefficient,and reconstruction time.Experimental results showed that our proposed DL-CS approach can significantly enhance signal quality compared with other compressed sampling methods.This is evidenced by a percentage root-mean-square difference,correlation coefficient,and reconstruction time results of 5%,0.9988,and 0.0108 s,respectively,obtained for sampling rates below 10%for the neutron ToF signal generated using an electron-beam-driven photoneutron source.The results showed that the proposed DL-CS approach significantly improves the signal quality compared with other compressed sampling methods,exhibiting excellent reconstruction accuracy and speed.
基金funded by the National Natural Science Foundation of China (Grant No.62203217)the Jiangsu Province Basic Research Program Natural Science Foundation (Grant No.BK20220885)+3 种基金the Hong Kong,Macao and Taiwan Science and Technology Cooperation Project of Special Foundation in Jiangsu Science and Technology Plan (Grant No.BZ2023057)the Fundamental Research Funds for the Central Universities (Grant No.NJ2024012)the China Postdoctoral Science Foundation (Grant No.GZC20242230)the Postgraduate Research & Practice Innovation Program of Jiangsu Province (Grant No.KYCX24_0586)。
文摘Cooperative task assignment is one of the key research focuses in the field of unmanned aerial vehicles(UAVs). In this paper, an energy learning hyper-heuristic(EL-HH) algorithm is proposed to address the cooperative task assignment problem of heterogeneous UAVs under complex constraints. First, a mathematical model is designed to define the scenario, complex constraints, and objective function of the problem. Then, the scheme encoding, the EL-HH strategy, multiple optimization operators, and the task sequence and time adjustment strategies are designed in the EL-HH algorithm. The scheme encoding is designed with three layers: task sequence, UAV sequence, and waiting time. The EL-HH strategy applies an energy learning method to adaptively adjust the energies of operators, thereby facilitating the selection and application of operators. Multiple optimization operators can update schemes in different ways, enabling the algorithm to fully explore the solution space. Afterward, the task order and time adjustment strategies are designed to adjust task order and insert waiting time. Through the iterative optimization process, a satisfactory assignment scheme is ultimately produced. Finally, simulation and experiment verify the effectiveness of the proposed algorithm.
基金Supported by National Natural Science Foundation of China(Grant No.51675040)。
文摘The electromagnetic pulse valve,as a key component in baghouse dust removal systems,plays a crucial role in the performance of the system.However,despite the promising results of intelligent fault diagnosis methods based on extensive data in diagnosing electromagnetic valves,real-world diagnostic scenarios still face numerous challenges.Collecting fault data for electromagnetic pulse valves is not only time-consuming but also costly,making it difficult to obtain sufficient fault data in advance,which poses challenges for small sample fault diagnosis.To address this issue,this paper proposes a fault diagnosis method for electromagnetic pulse valves based on deep transfer learning and simulated data.This method achieves effective transfer from simulated data to real data through four parameter transfer strategies,which combine parameter freezing and fine-tuning operations.Furthermore,this paper identifies a parameter transfer strategy that simultaneously fine-tunes the feature extractor and classifier,and introduces an attention mechanism to integrate fault features,thereby enhancing the correlation and information complementarity among multi-sensor data.The effectiveness of the proposed method is evaluated through two fault diagnosis cases under different operating conditions.In this study,small sample data accounted for 7.9%and 8.2%of the total dataset,and the experimental results showed transfer accuracies of 93.5%and 94.2%,respectively,validating the reliability and effectiveness of the method under small sample conditions.
