检索结果-南通市图书馆

ARGA-Unet:Advanced U-net segmentation model using residual grouped convolution and attention mechanism for brain tumor MRI image segmentation

在线全文

学校读者我要写书评

暂无评论

虚拟现实与智能硬件（中英文） 2024年第3期6卷 203-216页

作者： Siyi XUN Yan ZHANG Sixu DUAN Mingwei WANG jiangang chen Tong TONG Qinquan GAO Chantong LAM Menghan HU Tao TAN Faculty of Applied Sciences Macao Polytechnic UniversityMacao 999078China Jinan Branch of China Telecom Co.Ltd. Jinan 250000China School of Physics and Electronics Shandong Normal UniversityJinan 250000China Department of Dardiovascular Medicine Affiliated Hospital of Hangzhou Normal UniversityHangzhou 310000China Clinical School of Medicine Hangzhou Normal UniversityHangzhou 310000China Hangzhou Institute of Cardiovascular Diseases Hangzhou 310000China Shanghai Key Laboratory of Multidimensional Information Processing Shanghai 200000China School of Communication&Electronic Engineering East China Normal UniversityShanghai 200000China Engineering Research Center of Traditional Chinese Medicine Intelligent Rehabilitation Ministry of EducationShanghai 200000China College of Physics and Information Engineering Fuzhou UniversityFuzhou 350000China

Background Magnetic resonance imaging(MRI)has played an important role in the rapid growth of medical imaging diagnostic technology,especially in the diagnosis and treatment of brain tumors owing to its non invasive characteristics and superior soft tissue ***,brain tumors are characterized by high non uniformity and non-obvious boundaries in MRI images because of their invasive and highly heterogeneous *** addition,the labeling of tumor areas is time-consuming and *** To address these issues,this study uses a residual grouped convolution module,convolutional block attention module,and bilinear interpolation upsampling method to improve the classical segmentation network *** influence of network normalization,loss function,and network depth on segmentation performance is further *** In the experiments,the Dice score of the proposed segmentation model reached 97.581%,which is 12.438%higher than that of traditional U-net,demonstrating the effective segmentation of MRI brain tumor *** In conclusion,we use the improved U-net network to achieve a good segmentation effect of brain tumor MRI images.

关键词： Brain tumor MRI U-net Segmentation Attention mechanism Deep learning

Intelligent diagnosis of atrial septal defect in children using echocardiography with deep learning

在线全文

学校读者我要写书评

暂无评论

虚拟现实与智能硬件（中英文） 2024年第3期6卷 217-225页

作者： Yiman LIU Size HOU Xiaoxiang HAN Tongtong LIANG Menghan HU Xin WANG Wei GU Yuqi ZHANG Qingli LI jiangang chen Shanghai Key Laboratory of Multidimensional Information Processing School of Communication&Electronic EngineeringEast China Normal UniversityShanghaiChina Department of Pediatric Cardiology Shanghai Children's Medical CenterSchool of MedicineShanghai Jiao Tong UniversityShanghaiChina Department of Applied Mathematics Xi'an Jiaotong-Liverpool UniversitySchool of ScienceSuzhouChina School of Health Sciences and Engineering University of Shanghai for Science and TechnologyShanghaiChina Minhang District Centers for Disease Control and Prevention ShanghaiChina Faculty of Traditional Chinese Medicine Naval Medical UniversityShanghaiChina Engineering Research Center of Traditional Chinese Medicine Intelligent Rehabilitation Ministry of EducationShanghai 201203China

Background Atrial septal defect(ASD)is one of the most common congenital heart *** diagnosis of ASD via transthoracic echocardiography is subjective and *** The objective of this study was to evaluate the feasibility and accuracy of automatic detection of ASD in children based on color Doppler echocardiographic static images using end-to-end convolutional neural *** proposed depthwise separable convolution model identifies ASDs with static color Doppler images in a standard *** the standard views,we selected two echocardiographic views,i.e.,the subcostal sagittal view of the atrium septum and the low parasternal four-chamber *** developed ASD detection system was validated using a training set consisting of 396 echocardiographic images corresponding to 198 ***,an independent test dataset of 112 images corresponding to 56 cases was used,including 101 cases with ASDs and 153 cases with normal *** The average area under the receiver operating characteristic curve,recall,precision,specificity,F1-score,and accuracy of the proposed ASD detection model were 91.99,80.00,82.22,87.50,79.57,and 83.04,*** The proposed model can accurately and automatically identify ASD,providing a strong foundation for the intelligent diagnosis of congenital heart diseases.

关键词： Deep learning Atrial septal defect Echocardiography

高效MFe_(2)O_(4)(M=Zn、Mg、Cu和Mn)尖晶石催化剂应用于费托合成

在线全文

学校读者我要写书评

暂无评论

燃料化学学报（中英文） 2024年第5期52卷 667-676页

作者：王超陈建刚朱华青张文绍白洪彬张娟中国科学院山西煤炭化学研究所、煤炭高效低碳利用全国重点实验室山西太原030001 中国科学院大学北京100049 河南中托力合化学有限公司河南洛阳471002 北京三聚环保新材料股份有限公司北京100049

一系列尖晶石催化剂,包括ZnFe_(2)O_(4)、MgFe_(2)O_(4)、CuFe_(2)O_(4)和MnFe_(2)O_(4)被用于费托合成反应(Fischer-Tropsch synthesis,FTS)。Zn、Mg、Cu和Mn很容易与Fe形成尖晶石。其中,在前处理和反应过程中,Zn和Mg能够显著维持尖晶石结构,使得CO转化率较低。在反应过程中,Cu和Mn有利于碳化铁的生成,导致CuFe_(2)O_(4)和MnFe_(2)O_(4)对FTS性能影响显著。C_(2)^(=)-C_(4)^(=)ZnFe_(2)O_(4)对烃分布和C_(2)−C_(4)烯/烷比影响很小。MgFe_(2)O_(4)的C_(5+)选择性较低,同时由于Mg的碱性作用,从而提高了选择性和C_(2)−C_(4)烯/烷比。Cu可以促进催化剂的碳化,从而使CuFe_(2)O_(4)具有较高的活性。同时,CuFe_(2)O_(4)可以显著提高C_(5+)选C_(2)^(=)-C_(4)^(=)择性。此外,Cu可以促进H_(2)的解离和活化,从而有利于烯烃的二次加氢,降低选择性和C_(2)−C_(4)烯/烷比。虽然Mn在反应过程中会促进催化剂的碳化,但MnFe_(2)O_(4)对碳链的长短影响很小。然而,Mn能促进少量ε-Fe_(2)C的生成,这是导C_(2)^(=)-C_(4)^(=)致MnFe_(2)O_(4)具有较高选择性和C_(2)−C_(4)烯/烷比的原因。同时,所有尖晶石催化剂都具有较低的二氧化碳选择性,符合当前的绿色环保发展要求。

关键词：费托合成尖晶石物相结构 C_(2)-C_(4)烯/烷比

基于ArcGIS的地理微项目教学实践探索--以“浙江省人口分布”为例

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

地理教学 2024年第18期 27-32页

作者：陈坚刚蒋亚琴葛履龙浙江省春晖中学浙江绍兴312353 绍兴市教育教学研究院浙江绍兴312000 绍兴市高级中学浙江绍兴312000

信息技术的发展和应用是地理教学改革的助推器,对改变学生学习方式和教师教学方式具有重要作用。本文尝试将地理信息技术应用到微项目教学实践中,获取浙江省人口分布等数据,并利用ArcGIS软件对相关数据进行显示、分析等处理;并以此为基... 详细信息

信息技术的发展和应用是地理教学改革的助推器,对改变学生学习方式和教师教学方式具有重要作用。本文尝试将地理信息技术应用到微项目教学实践中,获取浙江省人口分布等数据,并利用ArcGIS软件对相关数据进行显示、分析等处理;并以此为基础,从提炼主题、设计任务、协作探究、科学评价四个维度,展开微项目教学,旨在培养学生的信息技术素养,提升地理实践力。

关键词：微项目教学地理信息技术 ArcGIS 地理实践力

Pollutant reduction effectiveness of low-impact development drainage system in a campus

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

Frontiers of Environmental Science & Engineering 2017年第4期11卷 113-120页

作者： Shuhan Zhang Yingying Meng Jiao Pan jiangang chen Beijing Water Science and Technology Institute Chegongzhuang West Road 21 Haidian District Beijing 100048 China

Building a rainwater system based on the idea of low-impact development （LID） is an important aspect of the current ＂sponge city＂ construction in China. The ＂sponge city＂ concept emphasizes that the runoff can permeate the soil or be stored temporarily, and rainwater could be used again when it is needed. Beijing is one of the earliest cities to study rainwater harvesting and LID techniques in *** long-terua monitoring of rainfall, runoff flow, and water quality of a campus demonstration project in Beijing, the runoff quantity and pollutant concentration variations have been analyzed. Furthermore, the runoffreduction effects of single LID measure, such as green roof, filtration chamber, and permeable pavement, have been investigated. Additionally, the overall reduction effectiveness of the LID system on the average annual rainfall runoff and pollution load has been discussed.

关键词： Low-impact development Rainwater harvesting Non-point-source pollution reduction Campus

气相色谱内标法对糟烧白酒中甲醇含量的检测与应用

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

酿酒 2024年第2期51卷 127-129页

作者：陈坚刚孙佳明会稽山绍兴酒股份有限公司浙江绍兴312000

建立气相色谱内标法,检测糟烧白酒中的甲醇含量,此方法精确度、灵敏度、回收率均良好;对半成品和成品糟烧白酒中的甲醇含量作了比较,发现半成品糟烧白酒中甲醇含量较高,需作进一步的抽样检测,以指导生产,改良工艺。

关键词：气相色谱内标法糟烧白酒甲醇含量指导生产

Coal Tar Derived Metallocene Catalyst for Polymerization of 1-Decene into Low-viscosity Poly-α-Olefin Lubricating Base Oil

在线全文

学校读者我要写书评

暂无评论

China Petroleum Processing & Petrochemical Technology 2022年第3期24卷 52-60页

作者： Wang Jiannan Liu Man chen Yilong Mohamed Abbas chen jiangang State Key Laboratory of Coal Conversion Institute of Coal ChemistryChinese Academy of SciencesTaiyuan 030001China University of Chinese Academy of Sciences Beijing 100049China State Key Laboratory of Biomass Thermal Chemistry Technology Wuhan 430223China Ceramics Department National Research CentreEI-Bohouth Str 12622CairoEgypt Faculty of Chemistry Jagiellonian UniversityUl.Gronostajowa 2Krakow30-387Poland

The production of poly-α-olefins(PAOs)has attracted attention due to their excellent viscosity-temperature dependence,wear characteristics,oxidative properties,and high thermal *** this study,indene extracted during coal tar refining was used as a raw material to synthesize a bis(indenyl)zirconium dichloride metallocene catalyst.A PAO with low viscosity and a high viscosity index was produced via the oligomerization of 1-decene in the presence of both the prepared metallocene and a methylaluminoxane(MAO)***,the effects of different synthesis reaction parameters,such as Al:Zr ratio,amount of catalyst,and reaction temperature,on the conversion ratio and product selectivity were investigated in *** produced PAO was thoroughly characterized using Fourier-transform infrared,^(13)C,and^(1)H nuclear magnetic resonance spectroscopies;gas chromatography;and viscosity *** 70℃,the metallocene catalyst created more stable active *** addition,the alkylation effect of MAO was ***,the obtained catalysis results demonstrated that a high conversion ratio of~93%was achieved at a low reaction temperature of 70℃,with a catalyst dosage of 0.0848 mmol and Al:Zr ratio of 8.48mmol:***,under these optimal conditions,the kinematic viscosity of PAO was 4.25 mm2/s at 100℃,and the viscosity index was 139,indicating good viscosity-temperature properties.

关键词： lubricant base oil poly-α-olefin coal tar metallocene catalyst polymerization

灰岩含水层定向钻孔注浆效果智能评价方法研究

在线全文

学校读者我要写书评

暂无评论

煤炭工程 2024年第1期56卷 63-69页

作者：李文昕贾东秀陈建刚傅子群陈军涛郭洪运刘磊山东科技大学能源与矿业工程学院山东青岛266590 山东能源新汶矿业集团有限责任公司邱集煤矿山东德州251105 山东能源新汶矿业集团有限责任公司榆树井煤矿内蒙古鄂尔多斯016299 煤炭资源高效开采与洁净利用国家重点实验室北京100013 山东科技大学矿业工程国家级实验教学示范中心山东青岛266590

基于黄河北煤田邱集煤矿现场注浆工程,通过XGboost、支持向量机、K近邻和BP人工神经网络四种算法,对灰岩定向钻孔的注浆效果进行科学分析。研究表明,XGboost、支持向量机、K近邻的模型精度均达不到0.9,而BP人工神经网络测试的拟合程度达... 详细信息

基于黄河北煤田邱集煤矿现场注浆工程,通过XGboost、支持向量机、K近邻和BP人工神经网络四种算法,对灰岩定向钻孔的注浆效果进行科学分析。研究表明,XGboost、支持向量机、K近邻的模型精度均达不到0.9,而BP人工神经网络测试的拟合程度达0.93,使用现场数据测试准确度达0.9,证明了BP人工神经网络评价注浆效果的可行性和准确性;最后运用MATLAB提出了一种基于BP人工神经网络的注浆效果智能化评价方法,并根据模型制作了简单的演示平台,实现了注浆效果评价的智能化与快捷化。

关键词：定向注浆效果评价机器学习 BP人工神经网络智能评价

在线全文

学校读者我要写书评

暂无评论

过滤对蒸馏杨梅酒甲醇含量的影响

酿酒 2024年第3期51卷 107-109页

作者：孙佳明陈坚刚郭柯洋会稽山绍兴酒股份有限公司浙江绍兴312032

采用气相色谱对发酵蒸馏制的杨梅酒中的甲醇含量进行测定分析,研究在杨梅酒发酵蒸馏过程中,过滤对蒸馏杨梅酒甲醇含量的影响,分析甲醇的主要来源是否由果肉中蒸馏出来的问题。结果表明在杨梅酒发酵的过程中,其甲醇很有可能富集在果肉当... 详细信息

采用气相色谱对发酵蒸馏制的杨梅酒中的甲醇含量进行测定分析,研究在杨梅酒发酵蒸馏过程中,过滤对蒸馏杨梅酒甲醇含量的影响,分析甲醇的主要来源是否由果肉中蒸馏出来的问题。结果表明在杨梅酒发酵的过程中,其甲醇很有可能富集在果肉当中,在蒸馏过程中被蒸馏出来。在杨梅酒蒸馏前对其进行过滤后再蒸馏测得其甲醇含量为0.28 g/L(折算成酒精100%vol),不进行过滤直接蒸馏其甲醇的含量为1.14 g/L(折算成酒精100%vol)。使用错流膜过滤掉果肉部分再进行蒸馏,能大幅度降低杨梅酒的甲醇含量。

关键词：气相色谱杨梅甲醇过滤