检索结果-南通市图书馆

基于麻雀搜索算法优化的BP神经网络模型对2型糖尿病肾病的预测研究

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

中国全科医学 2024年第8期27卷 961-970页

作者：邹琼吴曦张杨万毅陈长生空军军医大学军事预防医学系军队卫生统计学教研室、特殊作业环境危害评估与防治教育部重点实验室陕西省西安市710032 陕西中医药大学公共卫生学院陕西省咸阳市712046 空军军医大学卫勤训练基地陕西省西安市710032

背景糖尿病肾病(DN)是糖尿病常见的微血管并发症之一,发病率高,危害性大。早期发现DN对预防相关疾病非常重要。目前大多研究基于传统的统计预测方法,数据需满足其所要求的前提假设条件。近年来已无法很好满足其在DN预测领域的需求,有必... 详细信息

背景糖尿病肾病(DN)是糖尿病常见的微血管并发症之一,发病率高,危害性大。早期发现DN对预防相关疾病非常重要。目前大多研究基于传统的统计预测方法,数据需满足其所要求的前提假设条件。近年来已无法很好满足其在DN预测领域的需求,有必要尝试开展机器学习等新方法在DN预测领域的应用。目的利用LASSO回归和麻雀搜索算法(SSA)优化的BP神经网络(SSA-BP神经网络)构建DN预测模型。方法本研究时间为2023年4—8月,数据来源于公开的伊朗133例糖尿病患者的并发症数据。采用SPSS 26.0软件进行单因素分析,采用LASSO回归筛选变量。以是否患DN为因变量,分别用8∶2和7∶3的比例划分训练集和测试集,使用SSA-BP神经网络进行建模与分析,并与经典的机器学习模型对比预测性能以分析较优的DN模型。基于准确率、精确率、灵敏度、特异度、F1-score和受试者工作特征曲线下面积(AUC)指标进行模型评价。结果剔除9例1型糖尿病患者,本研究纳入的有效样本量为124例2型糖尿病(T2DM)患者,其中73例(58.9%)被诊断为DN患者。单因素分析显示年龄、BMI、糖尿病持续时间、空腹血糖(FBG)、糖化血红蛋白(HbA_(1c))、低密度脂蛋白(LDL)、高密度脂蛋白(HDL)、三酰甘油(TG)、收缩压(SBP)和舒张压(DBP)的T2DM患者DN危险因素(P<0.05)。训练集∶测试集=8∶2时,训练集(n=100)中有59例DN患者,测试集(n=24)含有14例DN患者。LASSO回归筛选出年龄、糖尿病持续时间、HbA_(1c)、LDL和SBP共5个影响因素。Logistic回归(LR)、K近邻(KNN)、支持向量机(SVM)、BP神经网络、SSA-BP神经网络模型在测试集的准确率分别为83.33%、79.17%、79.17%、87.50%、95.83%。F1-score分别为0.8462、0.8000、0.8000、0.8889、0.9600。训练集∶测试集=7∶3时,训练集(n=88)中有52例DN患者,测试集(n=36)含有21例DN患者。LASSO回归筛选出年龄、BMI、糖尿病持续时间、LDL、HDL、SBP和DBP这7个影响因素。LR、KNN、SVM、BP神经网络、SSA-BP神经网络模型在测试集的准确率分别为86.11%、86.11%、86.11%、72.22%、91.67%。F1-score分别为0.8718、0.8718、0.8649、0.7059、0.9091。结论LR、KNN和SVM模型在训练集∶测试集=7∶3时性能较好,BP神经网络和SSA-BP神经网络模型在训练集∶测试集=8∶2时性能较好。相较于BP神经网络模型和传统机器学习模型,SSA-BP神经网络模型的预测性能更佳,可及时准确识别T2DM DN患者,实现DN的早发现和早治疗,从而预防并减缓对其身体带来的危害。

关键词：糖尿病,2型糖尿病肾病神经网络,计算机预测模型

Role of substrate softness in stabilizing surface nanobubbles

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

Green Energy & Environment 2020年第3期5卷 374-380页

作者： changsheng chen Xianren Zhang Dapeng Cao State Key Laboratory of Organicinorganic Composites Beijing University of Chemical TechnologyBeijing100029China

The contact line pinning and supersaturation theory for the nanobubble stability has attracted extensive concerns from experimental investigators,and some experimenters argue that the contact line pinning is *** interpret the experimental observations,we have proposed previously through molecular dynamics simulations that the deformation of soft substrates caused by surface nanobubbles may play an important role in stabilizing surface nanobubbles,while yet no quantitative theory is available for explanation of this ***,the detailed mechanism of self-pinning-induced stability of surface nanobubbles is investigated through theoretical *** manipulating substrate softness,we find that the formation of surface nanobubbles may create a deformation ridge nearby their contact lines which leads to the self-pinning *** analysis shows that the formation of nanobubbles on sufficiently rigid substrates or on liquid-liquid interfaces corresponds to a local free energy maximum,while that on the substrates with intermediate softness corresponds to a local ***,the substrate softness could regulate the surface nanobubble *** critical condition for the self-pinning effect is determined based on contact line depinning,and the effect of gas supersaturation is ***,the approximate stability range for the surface nanobubbles is also predicted.

关键词： Surface nanobubble Soft substrate Stability Self-pinning

Interpreting the vulnerability of power systems in cascading failures using multi-graph convolutional networks

在线全文

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2022年第12期23卷 1848-1861页

作者： Supaporn LONAPALAWONG changsheng chen Can WANG Wei chen State Key Lab of CAD&CG Zhejiang UniversityHangzhou 310058China China Electric Power Research Institute Beijing 100192China College of Computer Science and Technology Zhejiang UniversityHangzhou 310058China

Analyzing the vulnerability of power systems in cascading failures is generally regarded as a challenging problem. Although existing studies can extract some critical rules, they fail to capture the complex subtleties under different operational conditions. In recent years, several deep learning methods have been applied to address this issue. However, most of the existing deep learning methods consider only the grid topology of a power system in terms of topological connections, but do not encompass a power system’s spatial information such as the electrical distance to increase the accuracy in the process of graph convolution. In this paper, we construct a novel power-weighted line graph that uses power system topology and spatial information to optimize the edge weight assignment of the line graph. Then we propose a multi-graph convolutional network(MGCN) based on a graph classification task, which preserves a power system’s spatial correlations and captures the relationships among physical components. Our model can better handle the problem with power systems that have parallel lines, where our method can maintain desirable accuracy in modeling systems with these extra topology features. To increase the interpretability of the model, we present the MGCN using layer-wise relevance propagation and quantify the contributing factors of model classification.

关键词： Power systems Vulnerability Cascading failures Multi-graph convolutional networks Weighted line graph

Diagnosis of the Forcing of Inertial-gravity Waves in a Severe Convection System

在线全文

学校读者我要写书评

暂无评论

Advances in Atmospheric Sciences 2016年第11期33卷 1271-1284页

作者： Lingkun RAN changsheng chen Institute of Atmospheric Physics Chinese Academy Sciences Beijing 100029 China Meteorological Bureau of Jilin Province Changchun 130062 China

The non-hydrostatic wave equation set in Cartesian coordinates is rearranged to gain insight into wave generation in a mesoscale severe convection system. The wave equation is characterized by a wave operator on the lhs, and forcing involving three terms—linear and nonlinear terms, and diabatic heating—on the rhs. The equation was applied to a case of severe convection that occurred in East China. The calculation with simulation data showed that the diabatic forcing and linear and nonlinear forcing presented large magnitude at different altitudes in the severe convection region. Further analysis revealed the diabatic forcing due to condensational latent heating had an important influence on the generation of gravity waves in the middle and lower levels. The linear forcing resulting from the Laplacian of potential-temperature linear forcing was dominant in the middle and upper levels. The nonlinear forcing was determined by the Laplacian of potential-temperature nonlinear forcing. Therefore, the forcing of gravity waves was closely associated with the thermodynamic processes in the severe convection case. The reason may be that, besides the vertical component of pressure gradient force, the vertical oscillation of atmospheric particles was dominated by the buoyancy for inertial gravity waves. The latent heating and potential-temperature linear and nonlinear forcing played an important role in the buoyancy tendency. Consequently, these thermodynamic elements influenced the evolution of inertial-gravity waves.

关键词： gravity convection forcing mesoscale vorticity perturbation latent Laplacian divergence heating

Stable bulk nanobubbles can be regarded as gaseous analogues of microemulsions

在线全文

学校读者我要写书评

暂无评论

Communications in Theoretical Physics 2023年第12期75卷 164-173页

作者： changsheng chen Hongguang Zhang Xianren Zhang State Key Laboratory of Organic–Inorganic Composites Beijing University of Chemical TechnologyBeijing 100029China

In our previous work[2022 ***.249685],we used molecular dynamics simulations to show that bulk nanobubbles can be stabilized by forming a compressed amphiphile monolayer at bubble *** observation closely matches the origin of stability of microemulsions and inspired us to propose here that,in certain cases,stable bulk nanobubbles can be regarded as gaseous analogues of microemulsions:the nanobubble phase and the bubble-containing solution phase coexist with the external gas *** three-phase coexistence is then validated by molecular dynamics *** stability mechanism for bulk nanobubbles is thus given:the formation of a compressed amphiphilic monolayer because of microbubble shrinking leads to a vanishing surface tension,and consequently the curvature energy of the monolayer dominates the thermodynamic stability of bulk *** the monolayer model,we further interpret several strange behaviors of bulk nanobubbles:gas supersaturation is not a prerequisite for nanobubble stability because of the vanishing surface tension,and the typical nanobubble size of 100 nm can be explained through the small bending constant of the ***,through analyzing the compressed amphiphile monolayer model we propose that bulk nanobubbles can exist ubiquitously in aqueous solutions.

关键词： bulk nanobubble stability compressed monolayer three-phase equilibrium

Impacts of oceanographic factors on interannual variability of the winter-spring cohort of neon flying squid abundance in the Northwest Pacific Ocean

在线全文

学校读者我要写书评

暂无评论

Acta Oceanologica Sinica 2017年第10期36卷 48-59页

作者： YU Wei chen Xinjun chen changsheng ZHANG Yu College of Marine Sciences Shanghai Ocean University National Engineering Research Center for Oceanic Fisheries Shanghai Ocean University Key Laboratory of Sustainable Exploitation of Oceanic Fisheries Resources of Ministry of Education Shanghai Ocean University Collaborative Innovation Center for Distant-water Fisheries School for Marine Science and Technology University of Massachusetts Dartmouth

The neon flying squid Ommastrephes bartramii is an economically important species in the Northwest Pacific Ocean. The life cycle of O. bartramii is highly susceptible to climatic and oceanic factors. In this study, we have examined the impacts of climate variability and local biophysical environments on the interannual variability of the abundance of the western winter-spring cohort of O. bartramii over the period of 1995–2011. The results showed that the squid had experienced alternant positive and negative Pacific Decadal Oscillation（PDO） over the past 17 years during which five El Ni？o and eight La Ni？a events occurred. The catch per unit effort（CPUE） was positively correlated with the PDO index（PDOI） at a one-year time lag. An abnormally warm temperature during the La Ni？a years over the positive PDO phase provided favorable oceanographic conditions for the habitats of ***, whereas a lower temperature on the fishing ground during the El Ni？o years over the negative PDO phase generally corresponded to a low CPUE. The same correlation was also found between CPUE and Chl a concentration anomaly. A possible explanation was proposed that the CPUE was likely related to the climateinduced variability of the large-scale circulation in the Northwest Pacific Ocean： high squid abundance often occurred in a year with a significant northward meander of the Kuroshio Current. The Kuroshio Current advected the warmer and food-rich waters into the fishing ground, and multiple meso-scale eddies arising from current instability enhanced the food retention on the fishing ground, all of which were favorable for the life stage development of the western squid stocks. Our results help better understand the potential process that the climatic and oceanographic factors affect the abundance of the winter-spring cohort of O. bartramii in the Northwest Pacific Ocean.

关键词： Ommastrephes bartramii squid abundance interannual variability PDO-CPUE relationship Kuroshio meandering meso-scale eddies

在线全文

学校读者我要写书评

暂无评论

基于文字边缘失真特征的翻拍图像篡改定位

中国科技论文 2024年第2期19卷 160-168,199页

作者：陈昌盛陈自炜李锡劲深圳大学电子与信息工程学院广东深圳518060

针对翻拍文档图像的篡改定位问题,提出一种基于文字边缘失真特征的翻拍图像篡改定位方法。从文字边缘分布、边缘梯度以及待检测文本与参考文本在边缘梯度上的差异3个方面构建了文字失真特征,并训练了一个基于深度神经网络的分类器进行... 详细信息

针对翻拍文档图像的篡改定位问题,提出一种基于文字边缘失真特征的翻拍图像篡改定位方法。从文字边缘分布、边缘梯度以及待检测文本与参考文本在边缘梯度上的差异3个方面构建了文字失真特征,并训练了一个基于深度神经网络的分类器进行决策。同时,为了评估检测方法的性能,构建了一个包含120张合法图像、1 200张翻拍篡改文档图像的数据集。实验结果表明:所提出的方法在跨库实验场景下词汇级别的ROC曲线下面积(area under ROC curve,AUC)和等错误率(equal error rate,EER)分别达到了0.84和0.23;与Forensic Similarity (128×128)和DenseFCN相比,所提出的特征结合LightDenseNet的方法在翻拍篡改文档数据集的跨库协议下,词汇级别的AUC指标分别提高了0.06和0.17。

关键词：文档图像翻拍攻击篡改定位文字边缘失真翻拍篡改文档数据库

在线全文

学校读者我要写书评

暂无评论

红毛菜与坛紫菜光合作用和品质的比较分析

应用海洋学学报 2024年第2期43卷 276-283页

作者：林密许凯徐燕纪德华王文磊陈昌生谢潮添集美大学水产学院福建厦门361021 福建省水产生物育种与健康养殖工程研究中心福建厦门361021 农业部东海海水健康与养殖重点实验室福建厦门361021

为研究两种经济红藻红毛菜(Bangia fucscopurpurea)和坛紫菜(Pyropia haitanensis)的光合作用和品质差异,选取一株野生型红毛菜CY和一株野生型坛紫菜NSD35作为实验材料,分别测定二者在相同培养条件下的关键光合参数和营养成分指标。结... 详细信息

为研究两种经济红藻红毛菜(Bangia fucscopurpurea)和坛紫菜(Pyropia haitanensis)的光合作用和品质差异,选取一株野生型红毛菜CY和一株野生型坛紫菜NSD35作为实验材料,分别测定二者在相同培养条件下的关键光合参数和营养成分指标。结果显示:红毛菜的最大电子传递速率、净光合速率、藻红蛋白含量分别比坛紫菜高29.6%、96.8%、60.5%;必需矿质元素钾、钙、镁、铁、锌的含量分别比坛紫菜高59.5%、39.5%、34.1%、86.2%、68.4%;红毛菜的必需氨基酸和风味氨基酸含量都高于坛紫菜。此外,二者的呼吸速率、最大光量子产率、叶绿素a、总糖、总蛋白和碘含量没有显著差异。研究结果表明,与坛紫菜NSD35相比,红毛菜CY的光合速率更快,矿质元素积累能力更强,必需氨基酸和风味氨基酸含量更高。研究结果有助于了解红毛菜和坛紫菜的光合作用和品质特点,为开发红毛菜种质资源提供数据参考。

关键词：海洋生物学红毛菜坛紫菜光合作用营养成分品质

Seasonal and Interannual Variability of Bering Strait Throughflow from AO-FVCOM and Observation

在线全文

学校读者我要写书评

暂无评论

Journal of Ocean University of China 2019年第3期18卷 615-625页

作者： DENG Yongfei GAO Guoping ZHANG Yu chen changsheng College of Marine Sciences Shanghai Ocean University Shanghai 201306 China College of Ocean Science and Engineering Shanghai Maritime University Shanghai 201306 China

A high-resolution Arctic Ocean-Finite Volume Community Ocean Model(AO-FVCOM) and observational current data from 14 mooring stations in Bering Strait and surrounding regions between 1990 and 2015 were used to study the seasonal and interannual variability of Bering Strait throughflow(BST). AO-FVCOM represented the BST with a climatological northward flux of 1.06 Sv, which was close to the observational mean of 0.94 ± 0.26 Sv. From the model results, the strongest volume flux was in summer, approximately 45% larger than that in winter. Interannual variability of BST was also indicated in the model results, and the maximum and minimum annual mean transports are in 2007 and 2012, respectively. AO-FVCOM showed larger differences from the observations in 2000, 2002, and 2015 than in other years, which may be related to the limitation of atmospheric forcing for the model. According to the driving mechanisms of BST, sea level difference(SLD) across the strait dominates the northward volume transport, and local wind is also important in forcing the seasonal variability of the BST and SLD patterns to change the BST indirectly.

关键词： AO-FVCOM Bering Strait throughflow seasonal and interannual variability in-situ observation local wind sea level difference