检索结果-南通市图书馆

A deep learning based fine-grained classification algorithm for grading of visual impairment in cataract patients

在线全文

学校读者我要写书评

暂无评论

Optoelectronics Letters 2024年第1期20卷 48-57页

作者： JIANG Jiewei ZHANG Yi XIE He YANG Jingshi GONG Jiamin LI Zhongwen School of Electronic Engineering Xi'an University of Posts and TelecommunicationsXi'an 710121China School of Ophthalmology and Optometry and Eye Hospital Wenzhou Medical UniversityWenzhou 325000China Ningbo Eye Hospital Wenzhou Medical UniversityNingbo 315000China

Recent advancements in artificial intelligence(AI)have shown promising potential for the automated screening and grading of cataracts.However,the different types of visual impairment caused by cataracts exhibit similar phenotypes,posing significant challenges for accurately assessing the severity of visual impairment.To address this issue,we propose a dense convolution combined with attention mechanism and multi-level classifier(DAMC_Net)for visual impairment grading.First,the double-attention mechanism is utilized to enable the DAMC_Net to focus on lesions-related regions.Then,a hierarchical multi-level classifier is constructed to enhance the recognition ability in distinguishing the severities of visual impairment,while maintaining a better screening rate for normal samples.In addition,a cost-sensitive method is applied to address the problem of higher false-negative rate caused by the imbalanced dataset.Experimental results demonstrated that the DAMC_Net outperformed ResNet50 and dense convolutional network 121(DenseNet121)models,with sensitivity improvements of 6.0%and 3.4%on the category of mild visual impairment caused by cataracts(MVICC),and 2.1%and 4.3%on the category of moderate to severe visual impairment caused by cataracts(MSVICC),respectively.The comparable performance on two external test datasets was achieved,further verifying the effectiveness and generalizability of the DAMC_Net.

关键词： visual classifier algorithm

Classification and reconstructive algorithm for nasal alar defect in Asians

在线全文

学校读者我要写书评

暂无评论

Chinese Journal of Plastic and Reconstructive Surgery 2024年第1期6卷 22-27页

作者： Renpeng Zhou Dongze Lyu Chen Wang Danru Wang Department of Plastic and Reconstructive Surgery Shanghai Ninth People's Hospital Shanghai Jiao Tong University School of Medicine

Background: The nasal alar defect in Asians remains a challenging issue, as do clear classification and algorithm guidance, despite numerous previously described surgical techniques. The aim of this study is to propose a surgical algorithm that addresses the appropriate surgical procedures for different types of nasal alar defects in Asian patients.Methods: A retrospective case note review was conducted on 32 patients with nasal alar defect who underwent reconstruction between 2008 and 2022. Based on careful analysis and our clinical experience, we proposed a classification system for nasal alar defects and presented a reconstructive algorithm. Patient data, including age,sex, diagnosis, surgical options, and complications, were assessed. The extent of surgical scar formation was evaluated using standard photography based on a 4-grade scar scale.Results: Among the 32 patients, there were 20 males and 12 females with nasal alar defects. The predominant cause of trauma in China was industrial factors. The majority of alar defects were classified as type I C(n=8,25%), comprising 18 cases(56.2%); there were 5 cases(15.6%) of type II defect, 7(21.9%) of type Ⅲ defect, and2(6.3%) of type IV defect. The most common surgical option was auricular composite graft(n=8, 25%), followed by bilobed flap(n=6, 18.8%), free auricular composite flap(n=4, 12.5%), and primary closure(n=3, 9.4%).Satisfactory improvements were observed postoperatively.Conclusion: Factors contributing to classifications were analyzed and defined, providing a framework for the proposed classification system. The reconstructive algorithm offers surgeons appropriate procedures for treating nasal alar defect in Asians.

关键词： Nasal alar Defect Classification algorithm Surgical methods

Metasurface-based computational imaging:a review

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Advanced Photonics 2024年第1期6卷 43-63页

作者： Xuemei Hu Weizhu Xu Qingbin Fan Tao Yue Feng Yan Yanqing Lu Ting Xu Nanjing University Collaborative Innovation Center of Advanced MicrostructuresNational Laboratory of Solid-State MicrostructuresNanjingChina Nanjing University School of Electronic Sciences and EngineeringNanjingChina Nanjing University College of Engineering and Applied SciencesJiangsu Key Laboratory of Artificial Functional MaterialsNanjingChina

Metasurface-based imaging has attracted considerable attention owing to its compactness,multifunctionality,and subwavelength coding capability.With the integration of computational imaging techniques,researchers have actively explored the extended capabilities of metasurfaces,enabling a wide range of imaging methods.We present an overview of the recent progress in metasurface-based imaging techniques,focusing on the perspective of computational imaging.Specifically,we categorize and review existing metasurface-based imaging into three main groups,including(i)conventional metasurface design employing canonical methods,(ii)computation introduced independently in either the imaging process or postprocessing,and(iii)an end-to-end computation-optimized imaging system based upon metasurfaces.We highlight the advantages and challenges associated with each computational metasurface-based imaging technique and discuss the potential and future prospects of the computational boosted metaimager.

关键词： metasurface computational imaging inverse problem algorithm

Intelligent Small Sample Defect Detection of Concrete Surface Using Novel Deep Learning Integrating Improved YOLOv5

在线全文

学校读者我要写书评

暂无评论

IEEE/CAA Journal of Automatica Sinica 2024年第2期11卷 545-547页

作者： Yongming Han Lei Wang Youqing Wang Zhiqiang Geng IEEE College of Information Science and Technology Beijing University of Chemical TechnologyBeijing 100029 Engineering Research Center of Intelligent PSE Ministry of Education in ChinaBeijing 100029China

Dear Editor,This letter presents an intelligent small sample defect detection of concrete surface using novel deep learning integrating the improved YOLOv5 based on the Wasserstein GAN(WGAN)enhancement algorithm.The proposed method is capable of producing top-notch data sets to address the issues of insufficient samples and substandard quality.

关键词： concrete integrating algorithm

Development of a Post Quantum Encryption Key Generation algorithm Using Electromagnetic Wave Propagation Theory

在线全文

学校读者我要写书评

暂无评论

Journal of Information Security 2024年第1期15卷 53-62页

作者： Vincent Mbonigaba Fulgence Nahayo Octave Moutsinga Okalas-Ossami Dieudonné Doctoral School and Faculty of Science and Technology University of Burundi Bujumbura Burundi LUMISTA-ISTA University of Burundi Bujumbura Burundi URMI Masuku University of Science and Technology Franceville Gabon

In today’s rapid widespread of digital technologies into all live aspects to enhance efficiency and productivity on the one hand and on the other hand ensure customer engagement, personal data counterfeiting has become a major concern for businesses and end-users. One solution to ensure data security is encryption, where keys are central. There is therefore a need to find robusts key generation implementation that is effective, inexpensive and non-invasive for protecting and preventing data counterfeiting. In this paper, we use the theory of electromagnetic wave propagation to generate encryption keys.

关键词： Key Wave Electromagnetic Cryptography Post Quantum Network Protocol Propagation algorithm

Attribute Reduction of Neighborhood Rough Set Based on Discernment

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Electronic Research and Application 2024年第1期8卷 80-85页

作者： Biqing Wang School of Mathematics and Computer Tongling UniversityTongling 244000Anhui ProvinceChina

For neighborhood rough set attribute reduction algorithms based on dependency degree,a neighborhood computation method incorporating attribute weight values and a neighborhood rough set attribute reduction algorithm using discernment as the heuristic information was proposed.The reduction algorithm comprehensively considers the dependency degree and neighborhood granulation degree of attributes,allowing for a more accurate measurement of the importance degrees of attributes.Example analyses and experimental results demonstrate the feasibility and effectiveness of the algorithm.

关键词： Neighborhood rough set Attribute reduction Discernment algorithm

‘Prediabetes’ as a practical distinctive window for workable fruitful wonders: Prevention and progression alert as advanced professionalism

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

World Journal of Clinical Pediatrics 2024年第1期13卷 1-6页

作者： Sunil Jain Department of Paediatrics Indian Armed Forces Medical Servicesc/o 56 APO 900244India

Diabetes is a devastating public health problem.Prediabetes is an intermediate stage in the disease processes leading to diabetes,including types 1 and 2 diabetes.In the article“Prediabetes in children and adolescents:An updated review,”the authors presented current evidence.We simplify and systematically clearly present the evidence and rationale for a conceptual framework we term the‘3ASs’:(1)Awareness Sensible;(2)algorithm Simple;and(3)Appealing Strategies.Policy makers and the public need to be alerted.The prevalence of prediabetes should send alarm bells ringing for parents,individuals,clinicians,and policy makers.Prediabetes is defined by the following criteria:impaired fasting glucose(100-125 mg/dL);impaired glucose tolerance(2 h postprandial glucose 140-199 mg/dL);or hemoglobin A1c values of 5.7%–6.4%.Any of the above positive test alerts for intervention.Clinical guidelines do not recommend prioritizing one test over the others for evaluation.Decisions should be made on the strengths and shortfalls of each test.Patient preferences and test accessibility should be taken into consideration.An algorithm based on age,physiological stage,health status,and risk factors is provided.Primordial prevention targeting populations aims to eliminate risk factors through public education and encouraging practices through environmental modifications.Access to healthy foods is provided.Primary prevention is for individuals with a prediabetes diagnosis and involves a structured program to reduce body weight and increase physical activity along with a healthy diet.An overall methodical move to a healthy lifestyle for lifelong health is urgently needed.Early energetic prediabetes action is necessary.

关键词： Obesity Overweight Awareness algorithm Lifestyle Physical exercise Screening Primordial prevention Primary prevention Adolescents

A New Method for Generating Conics

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Electronic Research and Application 2024年第3期8卷 125-128页

作者： Chengzhi Guo School of Mathematics and Statistics Qinghai Minzu UniversityXining 810007China

In this paper,based on the mean value theorem of differential,a new method of generating conics such as circles and parabolas is given,and the related algorithm for generating conics is designed.

关键词： Conic curve Differential Generate algorithm

Simplified algorithm for genetic subtyping in diffuse large B-cell lymphoma

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Signal Transduction and Targeted Therapy 2023年第5期8卷 2304-2314页

作者： Rong Shen Di Fu Lei Dong Mu-Chen Zhang Qing Shi Zi-Yang Shi Shu Cheng Li Wang Peng-Peng Xu Wei-Li Zhao Shanghai Institute of Hematology State Key Laboratory of Medical GenomicsNational Research Center for Translational Medicine at ShanghaiRuijin Hospital Affiliated to Shanghai Jiao Tong University School of MedicineShanghaiChina Department of Pathology Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of MedicineShanghaiChina Pôle de Recherches Sino-Français en Science du Vivant et Génomique Laboratory of Molecular PathologyShanghaiChina

Genetic classification helps to disclose molecular heterogeneity and therapeutic implications in diffuse large B-cell lymphoma(DLBCL).Using whole exome/genome sequencing,RNA-sequencing,and fluorescence in situ hybridization in 337 newly diagnosed DLBCL patients,we established a simplified 38-gene algorithm(termed‘LymphPlex’)based on the information on mutations of 35 genes and rearrangements of three genes(BCL2,BCL6,and MYC),identifying seven distinct genetic subtypes:TP53Mut(TP53 mutations),MCD-like(MYD88,CD79B,PIM1,MPEG1,BTG1,TBL1XR1,PRDM1,IRF4 mutations),BN2-like(BCL6 fusion,NOTCH2,CD70,DTX1,BTG2,TNFAIP3,CCND3 mutations),N1-like(NOTCH1 mutations),EZB-like(BCL2 fusion,EZH2,TNFRSF14,KMT2D,B2M,FAS,CREBBP,ARID1A,EP300,CIITA,STAT6,GNA13 mutations,with or without MYC rearrangement),and ST2-like(SGK1,TET2,SOCS1,DDX3X,ZFP36L1,DUSP2,STAT3,IRF8 mutations).Extended validation of 1001 DLBCL patients revealed clinical relevance and biological signature of each genetic subtype.TP53Mut subtype showed poor prognosis,characterized by p53 signaling dysregulation,immune deficiency,and PI3K activation.MCD-like subtype was associated with poor prognosis,activated B-cell(ABC)origin,BCL2/MYC double-expression,and NF-κB activation.BN2-like subtype showed favorable outcome within ABC-DLBCL and featured with NF-κB activation.N1-like and EZB-like subtypes were predominated by ABC-DLBCL and germinal center B-cell(GCB)-DLBCL,respectively.EZB-like-MYC+subtype was characterized by an immunosuppressive tumor microenvironment,while EZB-like-MYC-subtype by NOTCH activation.ST2-like subtype showed favorable outcome within GCB-DLBCL and featured with stromal-1 modulation.Genetic subtype-guided targeted agents achieved encouraging clinical response when combined with immunochemotherapy.Collectively,LymphPlex provided high efficacy and feasibility,representing a step forward to the mechanism-based targeted therapy in DLBCL.

关键词： lymphoma algorithm representing