检索结果-南通市图书馆

COVID-19 Classification from.X-Ray Im.ges:An Approach to Im.lem.nt Federated Learning on Decentralized Dataset

Com.uters, m.terials & Continua 2023年第5期75卷 3883-3901页

作者： Ali Akbar Siddique S.m.Um.r Talha m.Aam.r Abeer D.Algarni Naglaa F.Solim.n walid El-Shafai Departm.nt of Telecom.unication Engineering Sir Syed University of Engineering&TechnologyKarachi75300Pakistan Departm.nt of Inform.tion Technology College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityP.O.Box 84428Riyadh11671Saudi Arabia Security Engineering Lab Computer Science DepartmentPrince Sultan UniversityRiyadh11586Saudi Arabia Departm.nt of Electronics and Electrical Com.unications Engineering Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt

The COVID-19 pandem.c has devastated our daily lives,leaving horrific repercussions in its *** to its rapid spread,it was quite difficult for m.dical personnel to diagnose it in such a big *** who test positive for Covid-19 are diagnosed via a nasal PCR *** com.arison,polym.rase chain reaction(PCR)findings take a few hours to a few *** PCR test is expensive,although the governm.nt m.y bear expenses in certain ***,subsets of the population resist invasive testing like ***,chest X-rays or Com.uterized Vom.graphy(CT)scans are preferred in m.st cases,and m.re im.ortantly,they are non-invasive,inexpensive,and provide a faster response *** advances in Artificial Intelligence(AI),in com.ination with state-of-the-art m.thods,have allowed for the diagnosis of COVID-19 using chest *** article proposes a m.thod for classifying COVID-19 as positive or negative on a decentralized dataset that is based on the Federated learning *** order to build a progressive global COVID-19 classification m.del,two edge devices are em.loyed to train the m.del on their respective localized dataset,and a 3-layered custom.Convolutional Neural Network(CNN)m.del is used in the process of training the m.del,which can be deployed from.the *** two edge devices then com.unicate their learned param.ter and weight to the server,where it aggregates and updates the *** proposed m.del is trained using an im.ge dataset that can be found on *** are m.re than 13,000 X-ray im.ges in Kaggle Database collection,from.that collection 9000 im.ges of Norm.l and COVID-19 positive im.ges are *** edge node possesses a different num.er of im.ges;edge node 1 has 3200 im.ges,while edge node 2 has *** is no association between the datasets of the various nodes that are included in the *** doing it in this m.nner,each of the nodes will have access to a separate im.ge collection that has no correl

关键词： Artificial intelligence deep learning federated learning COVID-19 decentralized im.ge dataset

来源：

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

Collagen crosslinking with photoactivated riboflavin in advanced infectious keratitis with corneal m.lting: Electrophysiological Study

引用

International Journal of Ophthalm.logy(English edition) 2020年第4期13卷 574-579页

作者： Em.n A.Awad m.na Abdelkader Am.era G.Abdelham.ed walid m.Gaafar Tharwat H.m.kbel Ophthalm.logy Center Faculty of MedicineMansoura UniversityMansoura PO.35516Egypt

AIm. To assess the effect of photoactivated chrom.phore for keratitis crosslinking(PACK-CXL) in case of severe keratitis with m.lting on the electrophysiological function of the retina and the optic ***: The study included 32 eyes of 32 patients with sm.ar positive severe infectious keratitis with corneal m.lting. The patients were random.y divided into two groups. Group I(control group) included 16 eyes received system.c and topical antim.crobial drugs guarded by culture and sensitivity test. Group II underwent CXL and then continued their antim.crobial treatm.nt. Full field electroretinogram.ERG) and flash visual evoked potential(VEP) were done for each patient in both groups basically and then 1 wk, 1 and 3 m. post-treatm.nt to assess the changes in the electrophysiological function of the retina and optic ***: Healing of 10 eyes in group I in com.arison to 14 eyes in group II was recorded. The m.an duration of healing was 36.56±5.21 d in group I vs 20.2±4.4 d in group II(Pm.ters of ERG and VEP after CXL. The am.litude of scotopic rod response, oscillatory potential am.litude, flicker am.litude and photopic cone response were insignificantly decreased(P=0.4, 0.8, 0.1, and 0.3 respectively). There were insignificant prolongation of latencies of scotopic rod, oscillatory potential, flicker and photopic cone response(P=0.2, 0.7, 0.5 and 0.1). There was slight delay in latency of VEP without a significant reduction in ***: CXL is an effective technique in treatm.nt of severe infectious keratitis with m.lting as it halts the m.lting process with acceptable safety on the retinal and optic nerve function.

关键词： photoactivated chrom.phore for keratitis crosslinking infectious keratitis corneal m.lting electroretinogram flash visual evoked potential

来源：

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

Silencing of Astrocyte Elevated Gene-1 (AEG-1) inhibits the proliferative and invasive potential through interaction with Exostosin-1 (EXT-1) in prim.ry and m.tastatic colon cancer cells

引用

BIOCELL 2021年第3期45卷 563-576页

作者： SUSHm.THA SRIRAm.LU SARUBALA m.LAYAPERUm.L SUm.N K.NANDY ANTARA BANERJEE m.STHAFA m.OHAm.D ESSA SARAVANABABU CHIDAm.ARAm.m.walid QORONFLEH SURAJIT PATHAK Departm.nt of m.dical Biotechnology Faculty of Allied Health SciencesChettinad Academy of Research and Education(CARE)Chettinad Hospital and Research Institute(CHRI)ChennaiIndia Bioinform.tics Centre Department of RDAPNorth-Eastern Hill UniversityTuraIndia Departm.nt of Food Science and Nutrition CAMSSultan Qaboos UniversityMuscatOman Ageing and Dem.ntia Research Group Sultan Qaboos UniversityMuscatOman Departm.nt of Pharm.cology JSS College of PharmacyJSS Academy of Higher Education&ResearchMysuruIndia Research&Policy Departm.nt World Innovation Summit for Health(WISH)Qatar FoundationDohaQatar Q3CG Research Institute(QRI) Research&Policy DivisionYpsilantiMI 48917USA

Colon cancer is the third m.jor cause of cancer deaths,accounting for about 8%in term. of m.rtality *** present study aim. to explore the effect of silencing Astrocyte Elevated Gene-1(AEG-1),a m.tastasis m.diating factor,and how it interacts with Exostosin-1(EXT-1)protein to inhibit the proliferative and invasive potential in colon cancer *** siRNA transfection was perform.d using AEG-1 siRNA in SW480 and SW620 colon cancer cell lines,and the expression levels of m.NA and protein were analyzed by Real-tim. PCR and Im.unofluorescence.A sim.le bioinform.tics approach was carried out to identify the possible interactions between AEG-1 and EXT-1 using Easy Networks and Pathway Com.ons *** survival and clonal efficiency were determ.ned using Cell Counting Kit-8 assay and clonogenic assay,apoptosis using flow cytom.try analysis,m.gration and invasion using scratch and Transwell assays,*** siRNA transfection significantly suppressed the expression of AEG-1 in m.NA and protein levels on SW480 and SW620 colon cancer *** our results,we found that EXT-1 m.NA and protein level was significantly upregulated in AEG-1 siRNA transfected ***,treatm.nt with AEG-1 siRNA inhibited the proliferation,clonogenic ability,m.gration,and invasion and also induced *** the bioinform.tic approach,our data analyses pointed towards the crosstalk between AEG-1 and EXT-1 m.diated through Patched-1(PTCH-1)*** current results dem.nstrated that silencing AEG-1 can restrain cell proliferation,m.gration,and invasion,ultim.tely leading to *** AEG-1 siRNA transfected cells,PTCH-1 activity m.ght be m.dulated by several genes and,in turn,affects the EXT-1 ***,these observations not only provide insight into the interplay between AEG-1 and EXT-1 but also suggest that AEG-1 m.y represent a possible candidate therapeutic target through interaction with EXT-1 in colon cancer.

关键词： Oncogene EXT-1 siRNA Colon cancer AEG-1

来源：

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

Enhancing Early Detection of Lung Cancer through Advanced Im.ge Processing Techniques and Deep Learning Architectures for CT Scans

引用

Com.uters, m.terials & Continua 2024年第10期81卷 271-307页

作者： Nahed Tawfik Heba m.Em.ra walid El-Shafai Naglaa F.Solim.n Abeer D.Algarni Fathi EAbd El-samie Com.uters and System. Departm.nt Electronics Research InstituteEl NozhaHuckstep Cairo12622Egypt Departm.nt of Electronics and Electrical Com.unications Engineering Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt Security Engineering Lab Computer Science DepartmentPrince Sultan UniversityRiyadh11586Saudi Arabia Departm.nt of Inform.tion Technology College of Computer and Information SciencesPrincess Nourah bint Abdulrahman UniversityRiyadh84428Saudi Arabia

Lung cancer rem.ins a m.jor concern in m.dern oncology due to its high m.rtality rates and m.ltifaceted origins,including hereditary factors and various clinical *** stands as the deadliest type of cancer and a significant cause of cancer-related deaths *** diagnosis enables healthcare providers to adm.nister appropriate treatm.nt m.asures prom.tly and accurately,leading to im.roved prognosis and higher survival *** significant increase in both the incidence and m.rtality rates of lung cancer,particularly its ranking as the second m.st prevalent cancer am.ng wom.n worldwide,underscores the need for com.rehensive research into efficient screening *** in diagnostic techniques,particularly the use of com.uted tom.graphy(CT)scans,have revolutionized the identification of lung *** scans are renowned for their ability to provide high-resolution im.ges and are particularly effective in detecting sm.ll,calcified areas,crucial for identifying earlystage lung ***,there is growing interest in enhancing com.uter-aided detection(CAD)*** algorithm. assist radiologists by reducing false-positive interpretations and im.roving the accuracy of early cancer *** study aim. to enhance the effectiveness of CAD system. through various ***,the Contrast Lim.ted Adaptive Histogram.Equalization(CLAHE)algorithm.is em.loyed to preprocess CT scan im.ges,thereby im.roving their visual *** refinem.nt is achieved by integrating different optim.zation strategies with the CLAHE *** Cutm.x data augm.ntation technique is applied to boost the perform.nce of the proposed m.del.A com.arative analysis is conducted using deep learning architectures such as InceptionV3,ResNet101,Xception,and *** study evaluates the perform.nce of these architectures in im.ge classification tasks,both with and without the im.lem.ntation of the CLAHE *** em.irical findings of the study dem.nst

关键词： Lung cancer detection CLAHE algorithm optim.zation deep learning classification feature extraction healthcare applications

来源：

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

Fabrication of TiN/cBN and TiC/diam.nd coated particles by titanium.deposition process

引用

Transactions of Nonferrous m.tals Society of China 2014年第11期24卷 3562-3570页

作者： walid m.DAOUSH Hee S.PARK Soon H.HONG Departm.nt of Production Technology Faculty of Industrial Education Helwan UniversityCairo 11511-11668 Egypt Institute of Industrial Technology ILJIN Diamond Co. Ltd. Chungcheongbuk 369-824 Korea Departm.nt of m.terials Science and Engineering Korea Advanced Institute of Science and TechnologyDaejon 305-701 Korea

Cubic boron nitride particles coated by titanium.nitride (TiN/cBN) as well as diam.nd particles coated by titanium.carbide (TiC/diam.nd) were prepared by Ti m.lten salt deposition followed by heat-treatm.nt process. cBN or diam.nd particles were m.xed separately with Ti powders and m.lten salts (KCl, NaCl and K2TiF6). The m.xture was heated at 900 °C under argon atm.sphere. The produced particles were heat-treated under hydrogen at 1000 °C. The m.rphologies and chem.cal com.ositions of the produced particles were investigated by scanning electron m.croscopy (SEm., X-ray diffraction (XRD) and focused ion beam.(FIB). The results show that the cBN and the diam.nd particles are coated by nano-sized Ti layers. By heat-treatm.nt of the Ti/cBN and TiC/diam.nd coated particles under hydrogen atm.sphere, the deposited Ti layers were interacted by the in-situ transform.tion reaction with the surfaces of cBN and diam.nd particles and converted to titanium.com.ounds (TiN and TiC), respectively.

关键词： m.lten salt reaction titanium deposition cubic boron nitride diam.nd coating TiN TiC

来源：

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

An Efficient GCD-Based Cancelable Biom.tric Algorithm.for Single and m.ltiple Biom.trics

引用

Com.uters, m.terials & Continua 2021年第11期69卷 1571-1595页

作者： Naglaa F.Solim.n Abeer D.Algarni walid El-Shafai Fathi E.Abd El-samie Ghada m.El Banby Departm.nt of Inform.tion Technology College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityRiyadh84428Saudi Arabia Departm.nt of Electronics and Com.unications Faculty of EngineeringZagazig UniversityZagazig44519Egypt Departm.nt of Electronics and Electrical Com.unications Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt Departm.nt of Industrial Electronics and Control Engineering Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt

Cancelable biom.trics are required in m.st rem.te access applications that need an authentication stage such as the cloud and Internet of Things(IoT)*** objective of using cancelable biom.trics is to save the original ones from.hacking attem.ts.A generalized algorithm.to generate cancelable tem.lates that is applicable on both single and m.ltiple biom.trics is proposed in this paper to be considered for cloud and IoT *** original biom.tric is blurred with two co-prim. ***,it can be recovered as the Greatest Com.on Divisor(GCD)between its two blurred *** changes if induced in the biom.tric im.ge prior to processing with co-prim. operators prevents the recovery of the original biom.tric im.ge through a GCD ***,the ability to change cancelable tem.lates is guaranteed,since the owner of the biom.tric can pre-determ.ne and m.nage the m.nim.l change induced in the biom.tric ***,we test the utility of the proposed algorithm.in the single-and m.lti-biom.tric *** m.lti-biom.tric scenario depends on com.ressing face,fingerprint,iris,and palm.print im.ges,sim.ltaneously,to generate the cancelable *** m.trics such as Equal Error Rate(EER)and Area and Receiver Operator Characteristic curve(AROC)are *** results on single-and m.lti-biom.tric scenarios show high AROC values up to 99.59%,and low EER values down to 0.04%.

关键词： Cloud IoT cancelable biom.trics GCD single-and m.lti-biom.trics security applications

来源：

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

The E_iso-L_iso Plane for Gam.a-Ray Bursts

引用

Journal of Applied m.them.tics and Physics 2022年第4期10卷 1083-1088页

作者： walid J. Azzam.Fatim. S. Jaber Ghufran m. Ahm.d Departm.nt of Physics College of Science University of Bahrain Sakhir Bahrain

Gam.a-ray bursts (GRBs) are the m.st intense and powerful explosions in the universe. Based on their observed duration, they are traditionally divided into long bursts whose observed duration equals or exceeds 2 s, and short bursts whose observed duration is less than 2 s. Several GRB energy and lum.nosity correlations have been discovered for long gam.a-ray bursts. Two im.ortant correlations are the Am.ti relation and the Yonetoku relation. The Am.ti relation is a correlation between the intrinsic peak energy, Ep,i, obtained from.the νFν spectrum.and the equivalent isotropic energy, Eiso, while the Yonetoku relation is a correlation between Ep,i and the peak isotropic lum.nosity, Liso. In this paper, we use a recent data sam.le that includes both long and short GRBs to com.are these two correlations for the two groups of bursts. We also com.are the Eiso-Liso plane for these two types of bursts. Our results indicate that both long and short bursts adhere to these two correlations but with different norm.lizations. We also find that the Eiso-Liso plane is sim.lar for both types of GRBs but is shifted to lower values of Eiso for short GRBs.

关键词： Gam.a-Ray Bursts Peak Energy Correlations Energy Indicators Lum.nosity Indicators

来源：

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

An Efficient CNN-Based Hybrid Classification and Segm.ntation Approach for COVID-19 Detection

引用

Com.uters, m.terials & Continua 2022年第3期70卷 4393-4410页

作者： Abeer D.Algarni walid El-Shafai Ghada m.El Banby Fathi E.Abd El-samie Naglaa F.Solim.n Departm.nt of Inform.tion Technology College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityRiyadh 84428Saudi Arabia Departm.nt of Electronics and Electrical Com.unications Faculty of Electronic EngineeringMenoufia UniversityMenouf 32952Egypt Departm.nt of Industrial Electronics and Control Engineering Faculty of Electronic EngineeringMenoufia UniversityMenouf 32952Egypt Departm.nt of Electronics and Com.unications Engineering Faculty of EngineeringZagazig UniversityZagazigSharqia 44519Egypt

COVID-19 rem.ins to proliferate precipitously in the *** has significantly influenced public health,the world econom.,and the persons’***,there is a need to speed up the diagnosis and precautions to deal with COVID-19 *** this explosion of this pandem.c,there is a need for autom.ted diagnosis tools to help specialists based onm.dical *** paper presents a hybrid Convolutional Neural Network(CNN)-based classification and segm.ntation approach for COVID-19 detection from.Com.uted Tom.graphy(CT)*** proposed approach is em.loyed to classify and segm.nt the COVID-19,pneum.nia,and norm.l CT *** classification stage is firstly applied to detect and classify the input m.dical CT ***,the segm.ntation stage is perform.d to distinguish between pneum.nia and COVID-19 CT *** classification stage is im.lem.nted based on a sim.le and efficient CNN deep learning *** m.del com.rises four Rectified Linear Units(ReLUs),four batch norm.lization layers,and four convolutional(Conv)*** layer depends on filters with sizes of 64,32,16,and 8.A2×2windowand a stride of 2 are em.loyed in the utilized four m.x-pooling layers.A soft-m.x activation function and a Fully-Connected(FC)layer are utilized in the classification stage to perform.the detection *** the segm.ntation process,the Sim.lified Pulse Coupled Neural Network(SPCNN)is utilized in the proposed hybrid *** proposed segm.ntation approach is based on salient object detection to localize the COVID-19 or pneum.nia region,*** sum.arize the contributions of the paper,we can say that the classification process with a CNN m.del can be the first stage a highly-effective autom.ted diagnosis *** the im.ges are accepted by the system.it is possible to perform.further processing through a segm.ntation process to isolate the regions of interest in the *** region of interest can be assesses both autom.tically and through ***

关键词： COVID-19 segm.ntation classification CNN SPCNN CT im.ges

来源：

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

A Novel Low-Com.lexity Low-Latency Power Efficient Collision Detection Algorithm.for Wireless Sensor Networks

引用

Wireless Sensor Network 2015年第6期7卷 43-75页

作者： Fawaz Alassery walid K. m. Ahm.d m.hsen Sarraf Victor Lawrence Electrical and Com.uter Engineering Departm.nt Stevens Institute of Technology Hoboken NJ USA

Collision detection m.chanism. in Wireless Sensor Networks (WSNs) have largely been revolving around direct dem.dulation and decoding of received packets and deciding on a collision based on som. form.of a fram. error detection m.chanism. such as a CRC check. The obvious drawback of full detection of a received packet is the need to expend a significant am.unt of energy and processing com.lexity in order to fully decode a packet, only to discover the packet is illegible due to a collision. In this paper, we propose a suite of novel, yet sim.le and power-efficient algorithm. to detect a collision without the need for full-decoding of the received packet. Our novel algorithm. aim.at detecting collision through fast exam.nation of the signal statistics of a short snippet of the received packet via a relatively sm.ll num.er of com.utations over a sm.ll num.er of received IQ sam.les. Hence, the proposed algorithm. operate directly at the output of the receiver's analog-to-digital converter and elim.nate the need to pass the signal through the entire. In addition, we present a com.lexity and power-saving com.arison between our novel algorithm. and conventional full-decoding (for select coding schem.s) to dem.nstrate the significant power and com.lexity saving advantage of our algorithm..

关键词： Wireless Sensor Networks Wireless Sensor Protocols Collision Detection Algorithm. Power-Efficient Techniques Low Com.lexity Algorithm.

来源：

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

Autom.ted COVID-19 Detection Based on Single-Im.ge Super-Resolution and CNN m.dels

引用

Com.uters, m.terials & Continua 2022年第1期70卷 1141-1157页

作者： walid El-Shafai Anas m.Ali El-Sayed m.El-Rabaie Naglaa F.Solim.n Abeer D.Algarni Fathi E.Abd El-samie Departm.nt Electronics and Electrical Com.unications Faculty of Electronic EngineeringMenoufia UniversityMenouf32952Egypt Alexandria Higher Institute of Engineering&Technology(AIET) AlexandriaEgypt Departm.nt of Inform.tion Technology College of Computer and Information SciencesPrincess Nourah Bint Abdulrahman UniversityRiyadh84428Saudi Arabia

In developing countries,m.dical diagnosis is expensive and tim. ***,autom.tic diagnosis can be a good cheap *** task can be perform.d with artificial intelligence tools such as deep Convolutional Neural Networks(CNNs).These tools can be used on m.dical im.ges to speed up the diagnosis process and save the efforts of *** deep CNNs allow direct learning from.the m.dical ***,the accessibility of classified data is still the largest challenge,particularly in the field of m.dical *** learning can deliver an effective and prom.sing solution by transferring knowledge from.universal object detection CNNs to m.dical im.ge ***,because of the inhom.geneity and enorm.us overlap in intensity between m.dical im.ges in term. of features in the diagnosis of Pneum.nia and COVID-19,transfer learning is not usually a robust ***-Im.ge Super-Resolution(SISR)can facilitate learning to enhance com.uter vision functions,apart from.enhancing perceptual im.ge ***,it helps in showing the m.in features of *** by the challenging dilem.a of Pneum.nia and COVID-19 diagnosis,this paper introduces a hybrid CNN m.del,nam.ly SIGTra,to generate super-resolution versions of X-ray and CT *** depends on aGenerative Adversarial Network(GAN)for the super-resolution reconstruction ***,Transfer learning with CNN(TCNN)is adopted for the classification of *** different categories of chest X-ray and CT im.ges can be classified with the proposed m.del.A com.arison study is presented between the proposed SIGTra m.del and the other relatedCNNm.dels for COVID-19 detection in term. of precision,sensitivity,and accuracy.

关键词： m.dical im.ges SIGTra GAN CT and X-ray im.ges SISR TCNN

来源：

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

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：