The semimetal Bi has received increasing interest as an alternative to noble metals for use in plasmonic photocatalysis. To enhance the photocatalytic efficiency of metallic Bi, Bi microspheres modified by SiO2 nanopa...The semimetal Bi has received increasing interest as an alternative to noble metals for use in plasmonic photocatalysis. To enhance the photocatalytic efficiency of metallic Bi, Bi microspheres modified by SiO2 nanoparticles were fabricated by a facile method. Bi-O-Si bonds were formed between Bi and SiO2, and acted as a transportation channel for hot electrons. The SiO2@Bi microspheres exhibited an enhanced plasmon-mediated photocatalytic activity for the removal of NO in air under 280 nm light irradiation, as a result of the enlarged specific surface areas and the promotion of electron transfer via the Bi-O-Si bonds. The reaction mechanism of photocatalytic oxidation of NO by SiO2@Bi was revealed with electron spin resonance and in situ diffuse reflectance infrared Fourier transform spectroscopy experiments, and involved the chain reaction NO -> NO2 -> NO3- with center dot OH and center dot O-2(-) radicals as the main reactive species. The present work could provide new insights into the in-depth mechanistic understanding of Bi plasmonic photocatalysis and the design of high-performance Bi-based photocatalysts. (C) 2017, Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved.展开更多
脑电信号(electroencephalogram,EEG)在情感识别领域受到广泛关注。然而,现有方法多侧重于空间和时间维度,却忽视频段维度,存在时空频特征提取不足问题。提出一种基于多尺度卷积与时序建模的特征融合网络(multi-scale convolution and t...脑电信号(electroencephalogram,EEG)在情感识别领域受到广泛关注。然而,现有方法多侧重于空间和时间维度,却忽视频段维度,存在时空频特征提取不足问题。提出一种基于多尺度卷积与时序建模的特征融合网络(multi-scale convolution and temporal modeling for feature fusion network,MSCTF-Net)进行脑电情绪识别。将脑电信号重构为多维形式输入,设计多尺度空间卷积和频段卷积模块,从三维空频矩阵的空间和频段维度提取空间-频率信息;引入双向长短时记忆网络(Bi-LSTM)对二维时频矩阵进行时序建模,提取时间-频率特征;提出门控特征融合模块进行特征融合。模型在SEED数据集上的平均准确率为96.63%,在SEED-Ⅳ数据集上的平均准确率为91.58%,优于现有的多种深度学习方法。展开更多
目的探讨病种成本可视化分析系统的设计与构建。方法选择Power BI作为可视化分析工具。采集病种成本相关数据,导入Power BI并建立数据模型,编写度量值。采用Power BI Desktop设计可视化界面,对分析结果进行可视化展示,构建完整的病种成...目的探讨病种成本可视化分析系统的设计与构建。方法选择Power BI作为可视化分析工具。采集病种成本相关数据,导入Power BI并建立数据模型,编写度量值。采用Power BI Desktop设计可视化界面,对分析结果进行可视化展示,构建完整的病种成本可视化分析系统。结果可视化分析系统从总体概况、科室成本分析、DRG成本分析、DIP成本分析及病种成本分析五个角度进行分析展示,展示内容丰富,具有良好的洞察与分析效果。结论本研究应用Power BI快速实现了病种成本可视化分析系统的构建。病种成本可视化分析系统可为精细化管理和决策提供有效支持。展开更多
基金supported by the National Natural Science Foundation of China(21501016,51478070,21406022,21676037)the National Key R&D Project(2016YFC0204702)+4 种基金the Innovative Research Team of Chongqing(CXTDG201602014)the Natural Science Foundation of Chongqing(cstc2016jcyjA 0481,cstc2015jcyjA 0061)the Science and Technology Project of Chongqing Education Commission(KJ1600625,KJ1500637)the Application and Basic Science Project of Ministry of Transport of People's Republic of China(2015319814100)the Innovative Research Project from CTBU(yjscxx2016-060-36)~~
文摘The semimetal Bi has received increasing interest as an alternative to noble metals for use in plasmonic photocatalysis. To enhance the photocatalytic efficiency of metallic Bi, Bi microspheres modified by SiO2 nanoparticles were fabricated by a facile method. Bi-O-Si bonds were formed between Bi and SiO2, and acted as a transportation channel for hot electrons. The SiO2@Bi microspheres exhibited an enhanced plasmon-mediated photocatalytic activity for the removal of NO in air under 280 nm light irradiation, as a result of the enlarged specific surface areas and the promotion of electron transfer via the Bi-O-Si bonds. The reaction mechanism of photocatalytic oxidation of NO by SiO2@Bi was revealed with electron spin resonance and in situ diffuse reflectance infrared Fourier transform spectroscopy experiments, and involved the chain reaction NO -> NO2 -> NO3- with center dot OH and center dot O-2(-) radicals as the main reactive species. The present work could provide new insights into the in-depth mechanistic understanding of Bi plasmonic photocatalysis and the design of high-performance Bi-based photocatalysts. (C) 2017, Dalian Institute of Chemical Physics, Chinese Academy of Sciences. Published by Elsevier B.V. All rights reserved.
文摘脑电信号(electroencephalogram,EEG)在情感识别领域受到广泛关注。然而,现有方法多侧重于空间和时间维度,却忽视频段维度,存在时空频特征提取不足问题。提出一种基于多尺度卷积与时序建模的特征融合网络(multi-scale convolution and temporal modeling for feature fusion network,MSCTF-Net)进行脑电情绪识别。将脑电信号重构为多维形式输入,设计多尺度空间卷积和频段卷积模块,从三维空频矩阵的空间和频段维度提取空间-频率信息;引入双向长短时记忆网络(Bi-LSTM)对二维时频矩阵进行时序建模,提取时间-频率特征;提出门控特征融合模块进行特征融合。模型在SEED数据集上的平均准确率为96.63%,在SEED-Ⅳ数据集上的平均准确率为91.58%,优于现有的多种深度学习方法。
