检索结果-南通市图书馆

Early identification of stroke through deep learning with multi-modal human speech and movement data

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Neural Regeneration Research 2025年第1期20卷 234-241页

作者： Zijun Ou Haitao Wang Bin Zhang Haobang liang Bei Hu Longlong Ren Yanjuan Liu Yuhu Zhang Chengbo dai Hejun Wu Weifeng Li Xin Li School of Computer Science and Engineering Sun Yat-sen UniversityGuangzhouGuangdong ProvinceChina Department of Neurology Guangdong Neuroscience InstituteGuangdong Provincial People’s Hospital(Guangdong Academy of Medical Sciences)Southern Medical UniversityGuangzhouGuangdong ProvinceChina Department of Emergency Medicine Guangdong Provincial People’s Hospital(Guangdong Academy of Medical Sciences)Southern Medical UniversityGuangzhouGuangdong ProvinceChina

Early identification and treatment of stroke can greatly improve patient outcomes and quality of *** clinical tests such as the Cincinnati Pre-hospital Stroke Scale(CPSS)and the Face Arm Speech Test(FAST)are commonly used for stroke screening,accurate administration is dependent on specialized *** this study,we proposed a novel multimodal deep learning approach,based on the FAST,for assessing suspected stroke patients exhibiting symptoms such as limb weakness,facial paresis,and speech disorders in acute *** collected a dataset comprising videos and audio recordings of emergency room patients performing designated limb movements,facial expressions,and speech tests based on the *** compared the constructed deep learning model,which was designed to process multi-modal datasets,with six prior models that achieved good action classification performance,including the I3D,SlowFast,X3D,TPN,TimeSformer,and *** found that the findings of our deep learning model had a higher clinical value compared with the other ***,the multi-modal model outperformed its single-module variants,highlighting the benefit of utilizing multiple types of patient data,such as action videos and speech *** results indicate that a multi-modal deep learning model combined with the FAST could greatly improve the accuracy and sensitivity of early stroke identification of stroke,thus providing a practical and powerful tool for assessing stroke patients in an emergency clinical setting.

关键词： artificial intelligence deep learning diagnosis early detection FAST screening stroke

Research status and prospects of the fractal analysis of metal material surfaces and interfaces

在线全文

学校读者我要写书评

暂无评论

International Journal of Minerals,Metallurgy and Materials 2025年第1期32卷 20-38页

作者： Qinjin dai Xuefeng Liu Xin Ma Shaojie Tian Qinghe Cui Beijing Advanced Innovation Center for Materials Genome Engineering University of Science and Technology BeijingBeijing 100083China Beijing Laboratory of Metallic Materials and Processing for Modern Transportation University of Science and Technology BeijingBeijing 100083China Laboratory for Advanced Materials Processing of Ministry of Education University of Science and Technology BeijingBeijing 100083China School of Mechanical and Electrical Engineering Xinjiang Institute of EngineeringUrumqi 830023China

As a mathematical analysis method,fractal analysis can be used to quantitatively describe irregular shapes with self-similar or self-affine *** analysis has been used to characterize the shapes of metal materials at various scales and *** methods make it difficult to quantitatively describe the relationship between the regular characteristics and properties of metal material surfaces and ***,fractal analysis can be used to quantitatively describe the shape characteristics of metal materials and to establish the quantitative relationships between the shape characteristics and various properties of metal *** the perspective of two-dimensional planes and three-dimensional curved surfaces,this paper reviews the current research status of the fractal analysis of metal precipitate interfaces,metal grain boundary interfaces,metal-deposited film surfaces,metal fracture surfaces,metal machined surfaces,and metal wear *** relationship between the fractal dimensions and properties of metal material surfaces and interfaces is *** from three perspectives of fractal analysis,namely,research scope,image acquisition methods,and calculation methods,this paper identifies the direction of research on fractal analysis of metal material surfaces and interfaces that need to be *** is believed that revealing the deep influence mechanism between the fractal dimensions and properties of metal material surfaces and interfaces will be the key research direction of the fractal analysis of metal materials in the future.

关键词： metal material surfaces and interfaces fractal analysis fractal dimension homogeneity

SensFL:Privacy-Preserving Vertical Federated Learning with Sensitive Regularization

在线全文

学校读者我要写书评

暂无评论

Computer Modeling in Engineering & Sciences 2025年第1期142卷 385-404页

作者： Chongzhen Zhang Zhichen Liu Xiangrui Xu Fuqiang Hu Jiao dai Baigen Cai Wei Wang School of Automation and Intelligence Beijing Jiaotong UniversityBeijing100044China Shuohuang Railway Development Co. Ltd.National Energy GroupCangzhou062350China Beijing Key Laboratory of Security and Privacy in Intelligent Transportation School of Computer Science and TechnologyBeijing Jiaotong UniversityBeijing100044China

In the realm of Intelligent Railway Transportation Systems,effective multi-party collaboration is crucial due to concerns over privacy and data *** Federated Learning(VFL)has emerged as a promising approach to facilitate such collaboration,allowing diverse entities to collectively enhance machine learning models without the need to share sensitive training ***,existing works have highlighted VFL’s susceptibility to privacy inference attacks,where an honest but curious server could potentially reconstruct a client’s raw data from embeddings uploaded by the *** vulnerability poses a significant threat to VFL-based intelligent railway transportation *** this paper,we introduce SensFL,a novel privacy-enhancing method to against privacy inference attacks in ***,SensFL integrates regularization of the sensitivity of embeddings to the original data into the model training process,effectively limiting the information contained in shared *** reducing the sensitivity of embeddings to the original data,SensFL can effectively resist reverse privacy attacks and prevent the reconstruction of the original data from the *** experiments were conducted on four distinct datasets and three different models to demonstrate the efficacy of *** results show that SensFL can effectively mitigate privacy inference attacks while maintaining the accuracy of the primary learning *** results underscore SensFL’s potential to advance privacy protection technologies within VFL-based intelligent railway systems,addressing critical security concerns in collaborative learning environments.

关键词： Vertical federated learning privacy defenses

Elliptical inclusion in an anisotropic plane:non-uniform interface effects

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Applied Mathematics and Mechanics(English Edition) 2022年第5期43卷 667-688页

作者： Pengyu pei Ming dai School of Electrical and Mechanical Engineering Jinling Institute of TechnologyNanjing 211169China State Key Laboratory of Mechanics and Control of Mechanical Structures Nanjing University of Aeronautics and AstronauticsNanjing 210016China

We study the plane deformation of an elastic composite system made up of an anisotropic elliptical inclusion and an anisotropic foreign matrix surrounding the *** order to capture the influence of interface energy on the local elastic field as the size of the inclusion approaches the nanoscale,we refer to the Gurtin-Murdoch model of interface elasticity to describe the inclusion-matrix interface as an imaginary and extremely stiff but zero-thickness layer of a finite stretching *** opposed to isotropic cases in which the effects of interface elasticity are usually assumed to be uniform(described by a constant interface stretching modulus for the entire interface),the anisotropic case considered here necessitates non-uniform effects of interface elasticity(described by a non-constant interface stretching modulus),because the bulk surrounding the interface is *** this end,we treat the interface stretching modulus of the anisotropic composite system as a variable on the interface curve depending on the specific tangential direction of the *** then devise a unified analytic procedure to determine the full stress field in the inclusion and matrix,which is applicable to the arbitrary orientation and aspect ratio of the inclusion,an arbitrarily variable interface modulus,and an arbitrary uniform external loading applied *** non-uniform interface effects on the external loading-induced stress distribution near the interface are explored via a group of numerical *** is demonstrated that whether the nonuniformity of the interface effects has a significant effect on the stress field around the inclusion mainly depends on the direction of the external loading and the aspect ratio of the inclusion.

关键词： elliptical inclusion anisotropic material interface stress Gurtin-Murdoch model membrane-type interface

Nitrogen nutrition diagnosis for cotton under mulched drip irrigation using unmanned aerial vehicle multispectral images

在线全文

学校读者我要写书评

暂无评论

Journal of Integrative Agriculture 2023年第8期22卷 2536-2552页

作者： pei Sheng-zhao ZENG Hua-liang dai Yu-long BAI Wen-qiang FAN Jun-liang Key Laboratory of Agricultural Soil and Water Engineering in Arid and Semiarid Areas of Ministry of Education Northwest A&F UniversityYangling712100P.R.China

Remote sensing has been increasingly used for precision nitrogen management to assess the plant nitrogen status in a spatial and real-time *** nitrogen nutrition index(NNI)can quantitatively describe the nitrogen status of ***,the NNI diagnosis for cotton with unmanned aerial vehicle(UAV)multispectral images has not been evaluated *** study aimed to evaluate the performance of three machine learning models,i.e.,support vector machine(SVM),back propagation neural network(BPNN),and extreme gradient boosting(XGB)for predicting canopy nitrogen weight and NNI of cotton over the whole growing season from UAV *** results indicated that the models performed better when the top 15 vegetation indices were used as input variables based on their correlation ranking with nitrogen weight and *** XGB model performed the best among the three models in predicting nitrogen *** prediction accuracy of nitrogen weight at the upper half-leaf level(R^(2)=0.89,RMSE=0.68 g m^(-2),RE=14.62%for calibration and R^(2)=0.83,RMSE=1.08 g m^(-2),RE=19.71%for validation)was much better than that at the all-leaf level(R^(2)=0.73,RMSE=2.20 g m^(-2),RE=26.70%for calibration and R^(2)=0.70,RMSE=2.48 g m^(-2),RE=31.49%for validation)and at the plant level(R^(2)=0.66,RMSE=4.46 g m^(-2),RE=30.96%for calibration and R^(2)=0.63,RMSE=3.69 g m^(-2),RE=24.81%for validation).Similarly,the XGB model(R^(2)=0.65,RMSE=0.09,RE=8.59%for calibration and R^(2)=0.63,RMSE=0.09,RE=8.87%for validation)also outperformed the SVM model(R^(2)=0.62,RMSE=0.10,RE=7.92%for calibration and R^(2)=0.60,RMSE=0.09,RE=8.03%for validation)and BPNN model(R^(2)=0.64,RMSE=0.09,RE=9.24%for calibration and R^(2)=0.62,RMSE=0.09,RE=8.38%for validation)in predicting *** NNI predictive map generated from the optimal XGB model can intuitively diagnose the spatial distribution and dynamics of nitrogen nutrition in cotton fields,which can help farmers implement precise cotton nitrogen management in a t

关键词： UAV nitrogen diagnosis leaf nitrogen weight nitrogen nutrition index cotton

CircRNA: a rising star in plant biology

在线全文

学校读者我要写书评

暂无评论

Journal of Genetics and Genomics 2022年第12期49卷 1081-1092页

作者： pei Zhang Mingqiu dai National Key Laboratory of Crop Genetic Improvement Huazhong Agricultural UniversityWuhanHubei 430070China

Circular RNAs(circRNAs) are covalently closed single-stranded RNA molecules, which are widespread in eukaryotic cells. As regulatory molecules, circRNAs have various functions, such as regulating gene expression, binding mi RNAs or proteins, and being translated into proteins, which are important for cell proliferation and cell differentiation, individual growth and development, as well as many other biological processes. However, compared with that in animal models, studies of circRNAs in plants lags behind and,particularly, the regulatory mechanisms of biogenesis and molecular functions of plant circRNAs remain elusive. Recent studies have shown that circRNAs are wide spread in plants with tissue-or developmentspecific expression patterns and are responsive to a variety of environmental stresses. In this review, we summarize these advances, focusing on the regulatory mechanisms of biogenesis, molecular and biological functions of circRNAs, and the methods for investigating circRNAs. We also discuss the challenges and the prospects of plant circ RNA studies.

关键词： Plant circRNA circRNA biogenesis circRNA regulation Roles of circRNA Methodologies in circRNA studies

Allocating Fixed Costs with Considering the Return to Scale: A DEA Approach

在线全文

学校读者我要写书评

暂无评论

Journal of Systems Science & Complexity 2016年第5期29卷 1320-1341页

作者： dai Qianzhi LI Yongjun liang liang School of Management University of Science and Technology of China Institute of Policy and Management Chinese Academy of Sciences

When the allocated fixed cost is treated as the complement of other costs, conventional data envelopment analysis （DEA） researches have ignored the effect of the return to scale （RTS） in fixed cost Mlocation problems. This paper first demonstrates why the RTS should be considered in fixed cost allocation problems. Then treating the fixed cost as a complementary input, the authors investigate the relationship between the allocated cost and the variable return to scale （VRS） efficiency based on the super BCC DEA model. However, the infeasibility problem may exist in this situation. To deal with it, the authors propose an algorithm. The authors find that the super BCC efficiency is a monotone non-increasing function of the allocated cost. Based on the relationship, the authors finMly propose a fixed cost allocation approach in terms of principles as：（i） The fixed cost proportion allocated to inelastic DMUs should be consistent with their consumed cost proportion, and （ii） the same efficiency satisfaction degree to the rest DMUs. The optimal allocation scheme is unique. A numerical example and a real example of allocating fixed costs among 13 subsidiaries are employed to illustrate the proposed approach.

关键词： Data envelopment analysis (DEA) fixed cost allocation infeasibility return to scale(RTS).

Single Image Dehazing Based on Combining Dark Channel Prior and Scene Radiance Constraint

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Electronics 2016年第6期25卷 1114-1120页

作者： ZENG liang dai Yongzhen College of Computer National University of Defense Technology

Images captured in foggy or hazy weather conditions often suffer from poor visibility. The dark channel prior method has well solved the single image dehazing problem in nature, but it is invalid when the scene objects are inherently similar to the atmospheric light and no shadow is cast on them. We propose an efficient regularization method by adding a scene radiance constraint and combing the dark channel prior to remove hazes from a single input image. The experiments show that this improved algorithm can deal with various levels of foggy weather conditions, as well as greatly enhance the image's visibility and details. In addition, the recovered haze-free image has little or no halo artifacts.

关键词： Image dehazing Scene radiance constraint Dark channel prior Visibility enhancement Detail augment

Peripheral blood RNA biomarkers can predict lesion severity in degenerative cervical myelopathy

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Neural Regeneration Research 2025年第6期20卷 1764-1775页

作者： Zhenzhong Zheng Jialin Chen Jinghong Xu Bin Jiang Lei Li Yawei Li Yuliang dai Bing Wang Department of Spine Surgery The Second Xiangya Hospital of Central South UniversityChangshaHunan ProvinceChina

Degenerative cervical myelopathy is a common cause of spinal cord injury,with longer symptom duration and higher myelopathy severity indicating a worse *** numerous studies have investigated serological biomarkers for acute spinal cord injury,few studies have explored such biomarkers for diagnosing degenerative cervical *** study involved 30 patients with degenerative cervical myelopathy(51.3±7.3 years old,12 women and 18 men),seven healthy controls(25.7±1.7 years old,one woman and six men),and nine patients with cervical spondylotic radiculopathy(51.9±8.6 years old,three women and six men).Analysis of blood samples from the three groups showed clear differences in transcriptomic *** analysis identified 128 differentially expressed genes that were enriched in patients with neurological *** least absolute shrinkage and selection operator analysis,we constructed a five-gene model(TBCD,TPM2,PNKD,EIF4G2,and AP5Z1)to diagnose degenerative cervical myelopathy with an accuracy of 93.5%.One-gene models(TCAP and SDHA)identified mild and severe degenerative cervical myelopathy with accuracies of 83.3%and 76.7%,*** of two immune cell types(memory B cells and memory-activated CD4^(+)T cells)predicted levels of lesions in degenerative cervical myelopathy with 80%*** results suggest that peripheral blood RNA biomarkers could be used to predict lesion severity in degenerative cervical myelopathy.

关键词： biomarkers candidate genes degenerative cervical myelopathy gene expression analysis immune cell types neurological disabilities peripheral blood RNA profiles spinal cord injury