检索结果-南通市图书馆

Federated Learning with Privacy-preserving and Model IP-right-protection

在线全文

学校读者我要写书评

暂无评论

Machine Intelligence Research 2023年第1期20卷 19-37页

作者： Qiang Yang Anbu Huang Lixin Fan Chee Seng Chan Jian Han Lim Kam Woh ng Ding Sheng Ong bowen Li WeBank Shenzhen 518057China Hong Kong University of Science and Technology Hong Kong 999077China University of Malaya Kuala Lumpur 50603Malaysia University of Surrey Guildford GU27XHUK University of Aberystwyth Wales SY233DDUK Shanghai Jiao Tong University Shanghai 200240China

In the past decades,artificial intelligence(AI)has achieved unprecedented success,where statistical models become the central entity in ***,the centralized training and inference paradigm for building and using these models is facing more and more privacy and legal *** bridge the gap between data privacy and the need for data fusion,an emerging AI paradigm feder-ated learning(FL)has emerged as an approach for solving data silos and data privacy *** on secure distributed AI,feder-ated learning emphasizes data security throughout the lifecycle,which includes the following steps:data preprocessing,training,evalu-ation,and *** keeps data security by using methods,such as secure multi-party computation(MPC),differential privacy,and hardware solutions,to build and use distributed multiple-party machine-learning systems and statistical models over different data *** data privacy concerns,we argue that the concept of“model”matters,when developing and deploying federated models,they are easy to expose to various kinds of risks including plagiarism,illegal copy,and *** address these issues,we introduce FedIPR,a novel ownership verification scheme,by embedding watermarks into FL models to verify the ownership of FL models and protect model intellectual property rights(IPR or IP-right for short).While security is at the core of FL,there are still many articles re-ferred to distributed machine learning with no security guarantee as“federated learning”,which are not satisfied with the FL definition supposed to *** this end,in this paper,we reiterate the concept of federated learning and propose secure federated learning(SFL),where the ultimate goal is to build trustworthy and safe AI with strong privacy-preserving and *** provide a com-prehensive overview of existing works,including threats,attacks,and defenses in each phase of SFL from the lifecycle perspective.

关键词： Federated learning privacy-preserving machine learning security decentralized learning intellectual property protection

Deep learning Optical Character Recognition in PCB Dark Silk Recognition

在线全文

学校读者我要写书评

暂无评论

World Journal of Engineering and Technology 2023年第1期11卷 1-9页

作者： bowen Cai Shanghai Xiashu Intelligent Science Company Co. Ltd. Shanghai China

For Automatic Optical Inspection (AOI) machines that were introduced to Printed Circuit Board market more than five years ago, illumination technique and light devices are outdated. Images captured by old AOI machines are not easy to be recognized by typical optical character recognition (OCR) algorithms, especially for dark silk. How to effectively increase silk recognition accuracy is indispensable for improving overall production efficiency in SMT plant. This paper uses fine tuned Character Region Awareness for Text Detection (CRAFT) method to build model for dark silk recognition. CRAFT model consists of a structure similar to U-net, followed by VGG based convolutional neural network. Continuous two-dimensional Gaussian distribution was used for the annotation of image segmentation. CRAFT model is good at recognizing different types of printed characters with high accuracy and transferability. Results show that with the help of CRAFT model, accuracy for OK board is 95% (error rate is 5%), and accuracy for ng board is 100% (omission rate is 0%).

关键词： Deep Learning Dark Silk Computer Vision Pattern Recognition CRAFT Model Printed Circuit Board Electronics Manufacturing Services

PTEN在主动脉夹层病变组织中的表达变化及与血管平滑肌细胞收缩表型蛋白的关系

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

山东医药 2024年第14期64卷 27-30,35页

作者：刘恒娟王苏李博文闵兴平刘敏武汉大学人民医院麻醉科武汉430060 武汉大学人民医院心血管外科

目的观察磷酸酶和张力蛋白同源物(PTEN)在主动脉夹层(AD)患者及AD小鼠病变组织中的表达变化,并探讨其表达与血管平滑肌细胞收缩表型蛋白的关系。方法收集6例Stanford A型患者术中切除的病变主动脉(AD组)和6例供体心脏相应正常节段(对... 详细信息

目的观察磷酸酶和张力蛋白同源物(PTEN)在主动脉夹层(AD)患者及AD小鼠病变组织中的表达变化,并探讨其表达与血管平滑肌细胞收缩表型蛋白的关系。方法收集6例Stanford A型患者术中切除的病变主动脉(AD组)和6例供体心脏相应正常节段(对照组),采用HE染色观察主动脉结构,EVG染色观察主动脉弹力纤维,Western blotting法检测主动脉组织PTEN及血管收缩表型蛋白MYH11、α-SMA。另取3~4周龄C57BL/6J雄性小鼠12只,随机分为对照组和AD组各6只;取血管平滑肌细胞特异性敲除PTEN雄性小鼠6只作为PTENCKO组。AD组和PTENCKO组给予0.25%的β-氨基丙腈饲料喂养用于诱导AD,4周后给予血管紧张素Ⅱ1 000 ng/(kg·min)皮下泵注,连续2 d,构建AD模型;对照组给予正常饲料喂养,4周后给予等量生理盐水皮下泵注。处死小鼠,取主动脉组织,采用Western blotting法检测PTEN、MYH11、α-SMA及自噬相关蛋白LC3Ⅱ/Ⅰ、Beclin-1、p62。结果对照组血管壁组织层次清晰,细胞排列整齐,血管壁弹性纤维排列整齐;AD组血管壁结构层次不清,部分细胞丢失、排列不齐,血管壁弹性纤维崩解断裂,胶原沉积增多。与对照组相比,AD组主动脉组织PTEN蛋白表达升高,MYH11、α-SMA蛋白表达降低(P均2蛋白表达降低(P均2蛋白表达升高(P均<0.05)。结论 PTEN在AD患者及AD小鼠病变组织中高表达;抑制PTEN可减轻细胞自噬,恢复血管收缩表型蛋白表达,可能有助于缓解主动脉中层退行性变。

关键词：磷酸酶和张力蛋白同源物主动脉夹层主动脉中层退行性变细胞自噬

Fully Connected Convolutional Neural Network in PCB Soldering Point Inspection

同方期刊数据库博看期刊评论

在线全文

学校读者我要写书评

暂无评论

Journal of Computer and Communications 2022年第12期10卷 62-70页

作者： bowen Cai Shanghai Xiashu Intelligent Science Company Co. Ltd. Shanghai China

In Electronics Manufacturing Services (EMS) industry, Printed Circuit Board (PCB) inspection is tricky and hard, especially for soldering point inspection due to the extremely tiny size and inconsistent appearance for uneven heating in reflow soldering process. Conventional computer vision technique based on OpenCV or Halcon usually cause false positive call for originally good soldering point on PCB because OpenCV or Halcon use the pre-defined threshold in color proportion for deciding whether the specific soldering point is OK or ng (not good). However, soldering point forms are various after heating in reflow soldering process. This paper puts forward a VGG structure deep convolutional neural network, which is named SolderNet for processing soldering point after reflow heating process to effectively inspect soldering point status, reduce omission rate and error rate, and increase first pass rate. SolderNet consists of 11 hidden convolution layers and 3 densely connected layers. Accuracy reports are divided into OK point recognition and ng point recognition. For OK soldering point recognition, 92% is achieved. For ng soldering point recognition, 99% is achieved. The dataset is collected from KAGA Co. Ltd Plant in Suzhou. First pass rate at KAGA plant is increased from 25% to 80% in general.

关键词： Deep Learning Soldering Point Computer Vision Pattern Recognition Convolutional Neural Network Printed Circuit Board Electronics Manufacturing Services

Sensitivity and Performance of Cavity Optomechanical Field Sensors

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

Photonic Sensors 2012年第3期2卷 259-270页

作者： Stefan FORSTNER Joachim KNITTEL Eoin SHERIDAN Jon D. SWAIM Halina RUBINSZTEIN-DUNLOP Warwick P. bowen School of Mathematics and Physics University of Queensland St Lucia Queensland 4072 Australia

This article describes in detail a technique for model!ng cavity optomechanical field sensors. A magnetic or electric field induces a spatially varying stress across the sensor, which then induces a force on mechanical eigenmodes of the system. The force on each oscillator can then be determined from an overlap integral between magnetostrictive stress and the corresponding eigenmode, with the optomechanical coupling strength determining the ultimate resolution with which this force can be detected. Furthermore, an optomechanical magnetic field sensor is compared to other magnetic field sensors in terms of sensitivity and potential for miniaturization. It is shown that an optomechanical sensor can potentially outperform state-of-the-art magnetometers of similar size, in particular other sensors based on a magnetostrictive mechanism.

关键词： Cavity optomechanics magnetic field sensors magnetostriction integrated microcavity

艾曲泊帕乙醇胺片联合重组人血小板生成素注射液治疗原发免疫性血小板减少症患者的效果和安全性分析

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

黑龙江医学 2023年第14期47卷 1717-1720页

作者：安博文信阳市中心医院血液内科河南信阳464000

目的:分析艾曲泊帕乙醇胺片联合重组人血小板(PLT)生成素注射液对原发免疫性血小板减少症(ITP)患者PLT水平的影响。方法:选取2018年5月—2021年5月信阳市中心医院74例ITP患者作为研究对象,按随机数表法分为A组(n=37)和B组(n=37),B组予... 详细信息

目的:分析艾曲泊帕乙醇胺片联合重组人血小板(PLT)生成素注射液对原发免疫性血小板减少症(ITP)患者PLT水平的影响。方法:选取2018年5月—2021年5月信阳市中心医院74例ITP患者作为研究对象,按随机数表法分为A组(n=37)和B组(n=37),B组予以重组人PLT生成素注射液治疗,A组在B组基础上联合艾曲泊帕乙醇胺片治疗。对比两组患者免疫功能、PLT计数水平、起效时间、完全反应持续时间、PLT增幅、PLT输注数量、肝功能、肾功能[血清肌酐(Cr)、血尿素氮(BUN)]及不良反应发生率(腹泻、肌肉痉挛、头晕、发热)。结果:治疗后,A组PLT相关免疫球蛋白G(PAIgG)水平为(51.23±18.85) ng/107低于B组的(76.91±22.53) ng/107,差异有统计学意义(t=5.318,P0.05);治疗后,A组PLT水平为(130.07±35.94) 1.0×10^(9)/L高于B组的(101.12±30.25) 1.0×10^(9)/L,差异有统计学意义(t=3.749,P2±0.95) d、完全反应持续时间为(5.64±1.21) d短于B组的(5.41±1.82) d、(8.09±1.93) d,PLT增幅为(108.09±27.38)×10^(9)/L多于B组的(77.51±24.56)×10^(9)/L,PLT输注量为(77.51±24.56) U少于B组的(238.21±45.98) U,差异有统计学意义(t=7.081、6.542、5.057、10.752,P201,P>0.05);治疗后,两组患者Cr、BUN水平相比,差异无统计学意义(t=0.067、0.065,P>0.05);两组患者不良反应发生率对比,差异无统计学意义(χ^(2)=0.319,P>0.05)。结论:艾曲泊帕乙醇胺片与重组人PLT生成素共同治疗ITP患者效果确切,使PLT计数水平上升,加速症状恢复,还不损害肝肾功能,安全性更高。

关键词：血小板减少症艾曲泊帕乙醇胺片重组人血小板生成素血小板计数

在线全文

学校读者我要写书评

暂无评论

皮肤鳞癌及癌前病变中Foxp3蛋白的表达

重庆医科大学学报 2011年第6期36卷 660-663页

作者：刘干黄琨李惠重庆医科大学附属第一医院皮肤科重庆400016

目的:探讨正常皮肤和不同表皮肿瘤组织中叉头状转录因子p3(Forkhead box p3,Foxp3)的表达与临床意义。方法:应用免疫组化法检测15例日光性角化症(Actinic keratosis,AK),28例鲍温病(bowens’disease,BD),26例皮肤鳞癌(Squamous cellcarcinoma,SCC)和8例正常表皮组织(Normal epidermal tissue,NET)中Foxp3的表达。结果:Foxp3+调节性T细胞和Foxp3在皮肤癌前病变和恶性肿瘤中均有表达,在正常表皮组织中未见表达。其中,在肿瘤实质细胞中,Foxp3蛋白在AK、BD和SCC中的表达阳性率分别为46.7%、100%和100%。Foxp3蛋白在BD和SCC上的表达率均高于AK(P值均小于0.05),而在BD和SCC的表达率则无显著差异(P>0.05)。结论:Foxp3蛋白的表达在正常表皮、癌前病变、恶性肿瘤中的表达呈现逐渐增高的趋势,与表皮肿瘤的发展相关。

关键词：皮肤肿瘤叉头状转录因子p3 免疫逃逸

在线全文

学校读者我要写书评

暂无评论

甘薯愈伤组织分化及抗生素耐受性研究

中国农学通报 2009年第9期25卷 140-143页

作者：何博文岳昌武张义正四川大学生命科学学院四川省分子生物学与生物技术重点实验室成都610064

为研究不同甘薯品种愈伤组织对抗生素的耐受性,以及在不同激素配合使用时的分化情况,为甘薯转基因工程奠定基础,将不同浓度的羧苄青霉素、卡那霉素和潮霉素以及不同浓度和组合的NAA、BAP和2,4-D添加到MS培养基,接种不同来源的甘薯愈伤... 详细信息

为研究不同甘薯品种愈伤组织对抗生素的耐受性,以及在不同激素配合使用时的分化情况,为甘薯转基因工程奠定基础,将不同浓度的羧苄青霉素、卡那霉素和潮霉素以及不同浓度和组合的NAA、BAP和2,4-D添加到MS培养基,接种不同来源的甘薯愈伤组织并进行观察。不同甘薯品种对3种抗生素耐受性表现出一定差异,在较低羧苄青霉素浓度(100mg/L)的MS固体培养基中,所有品种愈伤组织生长良好。在较高羧苄青霉素浓度(≥500mg/L)的MS固体培养基中,南薯99和绵粉7号表现出较强的抗生素耐受性。NAA(0.1mg/L)与较高浓度BAP(≥1.0mg/L)配合使用,不同品种愈伤组织分化率为100%,较低浓度BAP(0.1mg/L)与不同浓度NAA配合使用,分化率极低。

关键词：甘薯抗生素耐受性愈伤组织分化

阿托伐他汀对自身免疫性脑脊髓炎小鼠髓鞘修复及RhoA/Rock1通路的影响

在线全文

学校读者我要写书评

暂无评论

中国比较医学杂志 2022年第4期32卷 68-76页

作者：张沁丽王玉芬刘红刘博雯长治医学院附属和平医院神经内科山西长治046000

目的探讨阿托伐他汀对自身免疫性脑脊髓炎(EAE)小鼠髓鞘修复及RhoA/Rock1通路的影响。方法采用MOG35-55免疫建立EAE小鼠模型,小鼠随机分为空白组、模型组、阿托伐他汀组、高脂饮食组、高脂饮食+阿托伐他汀组,每组6只,阿托伐他汀每只小... 详细信息

目的探讨阿托伐他汀对自身免疫性脑脊髓炎(EAE)小鼠髓鞘修复及RhoA/Rock1通路的影响。方法采用MOG35-55免疫建立EAE小鼠模型,小鼠随机分为空白组、模型组、阿托伐他汀组、高脂饮食组、高脂饮食+阿托伐他汀组,每组6只,阿托伐他汀每只小鼠每天按0.5 mL混悬液灌服,连续28 d。各组小鼠神经功能评分,采用苏木精-伊红(HE)、固蓝(LFB)染色、透射电镜及免疫组化染色方法检测各组小鼠脊髓组织炎症反应、髓鞘脱失及髓鞘再生情况;酶联免疫吸附法(ELISA)检测小鼠血清中肿瘤坏死因子-α(TNF-α)、白介素-6(IL-6)、一氧化氮(NO)的表达;蛋白免疫印迹(Western blot)法检测脑组织RAS同源基因家族成员A(RhoA)、Rho相关蛋白激酶1(Rock1)蛋白表达;实时荧光定量PCR(qRT-PCR)检测脊髓组织硫酸软骨素蛋白多糖(ng2)、髓鞘碱性蛋白(MBP)及脑组织RhoA、Rock1 mRNA表达。结果与空白组比较,模型组小鼠见较多炎性细胞浸润、发生明显脱髓鞘改变、部分髓鞘崩解、断裂、脱失;模型小鼠血清中TNF-α、IL-6、NO的含量及脑组织RhoA、Rock1蛋白和mRNA表达均明显升高(Png2、MBP蛋白及mRNA表达明显降低(Png2蛋白和mRNA表达(Png2 mRNA表达。结论阿托伐他汀能改善EAE小鼠炎性细胞浸润及脱髓鞘程度,降低高脂饮食EAE小鼠神经功能评分,其中作用机制可能与调节RhoA/Rock1通路改善脱髓鞘程度,从而发挥对EAE小鼠的治疗作用有关。

关键词：自身免疫性脑脊髓炎髓鞘修复阿托伐他汀 RhoA/ROCK-1通路