检索结果-南通市图书馆

Atherosclerotic plaque classif.cation in carotid ultrasound images using machine learning and explainable deep learning

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Intelligent Medicine 2024年第2期4卷 83-95页

作者： Soni Singh Pankaj K.Jain Neeraj Sharma Mausumi Pohit Sudipta roy School of.Vocational Studies and Applied Sciences Gautam Buddha UniversityGreater NoidaUttar PradeshIndia Artif.cial Intelligence and Data Science Jio Institute Navi MumbaiMaharashtraIndia School of.Biomedical Engineering Indian Institute of TechnologyVaranasiUttar PradeshIndia

Objective The incidence of.cardiovascular diseases(CVD)is rising rapidly *** f.rms of.CVD,such as stroke and heart attack,are more common among patients with certain *** development is a major f.ctor underlying cardiovascular events,such as heart attack and stroke,and its early detection may prevent such *** imaging of.carotid arteries is a usef.l method f.r diagnosis of.atherosclerotic plaques;however,an automated method to classif. atherosclerotic plaques f.r evaluation of.early-stage CVD is ***,we propose an automated method f.r classif.cation of.high-risk atherosclerotic plaque ultrasound *** f.ve deep learning(DL)models(VGG16,ResNet-50,GoogLeNet,XceptionNet,and SqueezeNet)were used f.r automated classif.cation and the results compared with those of.a machine learning(ML)-based technique,involving extraction of.23 texture f.atures f.om ultrasound images and classif.cation using a Support Vector Machine *** enhance model interpretability,output gradient-weighted convolutional activation maps(GradCAMs)were generated and overlayed on original *** A series of.indices,including accuracy,sensitivity,specif.city,f.-score,Cohen-kappa index,and area under the curve values,were calculated to evaluate model *** output images allowed visualization of.the most signif.cant ultrasound image *** GoogLeNet model yielded the highest accuracy(98.20%).Conclusion ML models may be also suitable f.r applications requiring low computational ***,DL models could be more completely automated than ML models.

关键词： Explainable deep learning Carotid artery Classif.cation VGG 16 ResNet-50 GoogLeNet XceptionNet SqueezeNet

Artif.cial Intelligence f.r Maximizing Agricultural Input Use Ef.iciency: Exploring Nutrient, Water and Weed Management Strategies

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Phyton-International Journal of.Experimental Botany 2024年第7期93卷 1569-1598页

作者： Sumit Sow Shivani Ranjan Mahmoud f.Seleiman Hiba M.Alkharabsheh Mukesh Kumar Navnit Kumar Smruti Ranjan Padhan Dhirendra Kumar roy Dibyajyoti Nath Harun Gitari Daniel O.Wasonga Department of.Agronomy Dr.Rajendra Prasad Central Agricultural UniversityPusaSamastipurBihar848125India Department of.Plant Production College of Food and Agriculture SciencesKing Saud UniversityRiyadh11451Saudi Arabia Department of.Crop Sciences Faculty of AgricultureMenoufia UniversityShibin El-Kom32514Egypt Department of.Water Resources and Environmental Management Faculty of Agricultural TechnologyAl Balqa Applied UniversitySalt19117Jordan Division of.Agronomy ICAR-Indian Agricultural Research InstitutePusa CampusNew Delhi110012India Department of.Soil Science Dr.Rajendra Prasad Central Agricultural UniversityPusaSamastipurBihar848125India Department of.Agricultural Science and Technology School of Agriculture and Environmental SciencesKenyatta UniversityNairobi43844Kenya Department of.Crop Sciences University of Illinois Urbana-ChampaignUrbanaIL 61801USA

Agriculture plays a crucial role in the economy,and there is an increasing global emphasis on automating agri-cultural *** the tremendous increase in population,the demand f.r f.od and employment has also increased signiﬁ*** methods traditionally used to meet these requirements are no longer ade-quate,requiring solutions to issues such as excessive herbicide use and the use of.chemical *** of.technologies such as the Internet of.Things,wireless communication,machine learning,artiﬁcial intelligence(AI),and deep learning shows promise in addressing these ***,there is a lack of.comprehensive documentation on the application and potential of.AI in improving agricultural input ef.*** address this gap,a desk research approach was used by utilizing peer-reviewed electronic databases like PubMed,Scopus,Goo-gle Scholar,Web of.Science,and Science Direct f.r relevant *** of.327 initially identiﬁed articles,180 were deemed pertinent,f.cusing primarily on AI’s potential in enhancing yield through better management of.nutrients,water,and *** into account researchﬁndings worldwide,we f.und that AI technologies could assist f.rmers by providing recommendations on the optimal nutrients to enhance soil quality and deter-mine the best time f.r irrigation or herbicide *** present status of.AI-driven automation in agricul-ture holds signiﬁcant promise f.r optimizing agricultural input utilization and reducing resource waste,particularly in the context of.three pillars of.crop management,i.e.,nutrient,irrigation,and weed management.

关键词： Agriculture artiﬁcial intelligence crop management nutrient irrigation weed management resource use ef.ciency

Quantized Redshif. and its Signif.cance f.r Recent Observation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Research in Astronomy and Astrophysics 2024年第9期24卷 158-168页

作者： Arindam Mal Sarbani Palit Christopher C.f.lton Sisir roy Indian Space Research Organization Space Applications Centre Indian Statistical Institute CVPR Unit Protostar Inc. 15652 Carrousel DriveCanyon Country National Institute of.Advanced Studies IISC Campus

With the recent observational evidence in extragalactic astronomy, the interpretation of.the nature of.quasar redshif. continues to be a research interest. Very high redshif.s are being detected f.r extragalactic objects that are presumably very distant and young while also exhibiting properties that are characteristic of.a more mature galaxy such as ours. According to Halton Arp and Geof.rey Burbidge, redshif. disparities consist of.an intrinsic component and are related to an evolutionary process. Karlsson observed redshif. periodicity at integer multiples of..089 in log scale and Burbidge observed redshif. periodicity at integer multiples of.0.061 in linear scale. Since Singular Value Decomposition based periodicity estimation is known to be superior f.r noisy data sets, especially when the data contain multiple harmonics and overtones, mainly irregular in nature, we have chosen it to be our primary tool f.r analysis of.the quasar-galaxy pair redshif. data. We have observed a f.ndamental periodicity of..051 with a conf.dence interval of.95%in linear scale with the site-available Sloan Digital Sky Survey Data Release 7 (SDSS DR7) quasar-galaxy pair data set. We have independently generated quasar-galaxy pair data sets f.om both 2d f.and SDSS and f.und f.ndamental periodicities of.0.077 and 0.089, respectively, in log scale with a conf.dence interval of.95%.

关键词： galaxies: distances and redshif.s (galaxies:) quasars: absorption lines cosmology: observations PHYSICAL ASSOCIATION QUASAR PERIODICITY GALAXIES RADIO SDSS

Development of.a Lightweight Model f.r Handwritten Dataset Recognition: Bangladeshi City Names in Bangla Script

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第8期80卷 2633-2656页

作者： MdMahbubur Rahman Tusher f.hmid Al f.rid MdAl-Hasan Abu Saleh Musa Miah Susmita roy Rinky Mehedi Hasan Jim Sarina Mansor MdAbdur Rahim Hezerul Abdul Karim Department of.Computer Science and Engineering Bangladesh Army University of Science and Technology(BAUST)Saidpur5310Bangladesh f.culty of.Engineering Multimedia UniversityCyberjaya63100Malaysia Pabna University of.Science and Technology Pabna6600Bangladesh

The context of.recognizing handwritten city names,this research addresses the challenges posed by the manual inscription of.Bangladeshi city names in the Bangla *** today’s technology-driven era,where precise tools f.r reading handwritten text are essential,this study f.cuses on leveraging deep learning to understand the intricacies of.Bangla *** existing dearth of.dedicated datasets has impeded the progress of.Bangla handwritten city name recognition systems,particularly in critical areas such as postal automation and document ***,no prior research has specif.cally targeted the unique needs of.Bangla handwritten city name *** bridge this gap,the study collects real-world images f.om diverse sources to construct a comprehensive dataset f.r Bangla Hand Written City name *** emphasis on practical data f.r system training enhances *** research f.rther conducts a comparative analysis,pitting state-of.the-art(SOTA)deep learning models,including Ef.icientNetB0,VGG16,ResNet50,DenseNet201,InceptionV3,and Xception,against a custom Convolutional Neural Networks(CNN)model named“Our CNN.”The results showcase the superior perf.rmance of.Our CNN,”with a test accuracy of.99.97% and an outstanding f. score of.99.95%.These metrics underscore its potential f.r automating city name recognition,particularly in postal *** study concludes by highlighting the signif.cance of.meticulous dataset curation and the promising outlook f.r custom CNN *** encourages f.ture research avenues,including dataset expansion,algorithm ref.nement,exploration of.recurrent neural networks and attention mechanisms,real-world deployment of.models,and extension to other regional languages and *** recommendations of.er exciting possibilities f.r advancing the f.eld of.handwritten recognition technology and hold practical implications f.r enhancing global postal services.

关键词： Handwritten recognition Bangladeshi city names Bangla handwritten city name automated postal services

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

用预热解决感应淬火工件的尺寸稳定性问题

铸锻热（热处理实践） 1993年第3期 15-17页

作者： roy f. KERN 谢燮揆

感应淬火前进行预热可以增加残余压应力,改善淬硬层形态,并获得最佳的显微组织,从而能解决许多问题。在感应淬火履带滚轮时。

关键词：感应热处理淬火尺寸稳定性

维普期刊数据库

在线全文

学校读者我要写书评

暂无评论

眼症状的鉴别诊断

Credit Card f.aud Detection Using Machine Learning Techniques

1981年

作者： f.H.roy 吴乐正译

来源：汇雅图书（南通市图书馆... 评论

在线全文

汇雅图书（南通市图书馆）

学校读者我要写书评

暂无评论

Journal of.Computer and Communications 2024年第6期12卷 1-11页

作者： Ananya Sarker Must. Asma Yasmin Md. Atikur Rahman Md. Harun Or Rashid Bristi Rani roy Department of.Computer Science and Engineering Bangladesh Army University of Engineering & Technology (BAUET) Natore Bangladesh

Credit card companies must be able to identif. f.audulent credit card transactions so that clients are not charged f.r items they did not purchase. Previously, many machine learning approaches and classif.ers were used to detect f.audulent transactions. However, because f.aud patterns are always changing, it is becoming increasingly vital to investigate new f.auds and develop the model based on the new patterns. The purpose of.this research is to create a machine learning classif.er that not only detects f.aud but also detects legitimate transactions. As a result, the model should have excellent accuracy, precision, recall, and f.-score. As a result, we began with a large dataset in this study and used f.ur machine learning classif.ers: Support Vector Machine (SVM), Decision Tree, Naïve Bayes, and Random f.rest. The random f.rest classif.er scored 99.96% overall accuracy with the best precision, recall, f.-score, and Matthews correlation coef.icient in the experiments.

关键词： Support Vector Machine Decision Tree Nave Bayes Random f.rest Matthews Correlation

儿童青少年体力活动与骨健康研究的10大问题

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

北京体育大学学报 2016年第1期39卷 46-52,60页

作者： Kathleen f.Janz David Q.Thomas M.allison f.rd Skip M.Williams 爱荷华大学 Iowa CityIA 52242USA 伊利诺斯州立大学 NornmlIL 61790USA 密西西比大学 OxfordMS 38677USA

证据强烈支持体力活动对骨强度有积极的因果效应,并指出儿童进行体力活动对预防骨质疏松有长远益处。年轻时的健康骨骼可能对预防成年后由于骨量丢失而发生骨质疏松有重要作用。家庭、学校(特别是体育课)和社区是进行骨骼健康促进体力... 详细信息

证据强烈支持体力活动对骨强度有积极的因果效应,并指出儿童进行体力活动对预防骨质疏松有长远益处。年轻时的健康骨骼可能对预防成年后由于骨量丢失而发生骨质疏松有重要作用。家庭、学校(特别是体育课)和社区是进行骨骼健康促进体力活动的重要场所。然而,有关健康促进和体育课对骨骼健康的话题很少,因此最佳的实践方法还不清楚。基于文献资料的了解,我们提出了应该被解决的健康促进和体育课研究的10大问题,从而促进儿童青少年的体力活动。

关键词：儿童时期运动机械(冲击)载荷骨骼健康

维普期刊数据库

制作高速GaAs场效应晶体管用的微细光刻技术

在线全文

学校读者我要写书评

暂无评论

微纳电子技术 1987年第1期 46-54页

作者： J.f.allison 江泽流

本文论述了用来制作高速GaAs金属-半导体场效应晶体管(MESf.T)的源、漏和栅电极的微细光刻技术。用剥离法和常规光刻制备了MESf.T的源和漏电极,并叙述了用斜角蒸发形成亚微米栅长的工艺过程。还详述了用电子束刻蚀技术进行栅极的描绘乖... 详细信息

本文论述了用来制作高速GaAs金属-半导体场效应晶体管(MESf.T)的源、漏和栅电极的微细光刻技术。用剥离法和常规光刻制备了MESf.T的源和漏电极,并叙述了用斜角蒸发形成亚微米栅长的工艺过程。还详述了用电子束刻蚀技术进行栅极的描绘乖抗蚀剂剖面的控制。提供了在制作GaAs场效应晶体管(f.T)和微波单片集成电路(MMIC)过程中,应用这些工艺技术实例的有关数据。

关键词：场效应晶体管单极晶体管场效应器件抗蚀剂 GaAs 光刻技术栅长 MESf.T 栅极栅电极钻蚀电子束光刻电子束加工光学光刻欧姆接触