Damage detection in structures is performed via vibra-tion based structural identification. Modal information, such as fre-quencies and mode shapes, are widely used for structural dama-ge detection to indicate the hea...Damage detection in structures is performed via vibra-tion based structural identification. Modal information, such as fre-quencies and mode shapes, are widely used for structural dama-ge detection to indicate the health conditions of civil structures.The deep learning algorithm that works on a multiple layer neuralnetwork model termed as deep autoencoder is proposed to learnthe relationship between the modal information and structural stiff-ness parameters. This is achieved via dimension reduction of themodal information feature and a non-linear regression against thestructural stiffness parameters. Numerical tests on a symmetri-cal steel frame model are conducted to generate the data for thetraining and validation, and to demonstrate the efficiency of theproposed approach for vibration based structural damage detec-tion.展开更多
基于运动想象的脑机接口系统一直是海内外研究学者的关注对象。针对传统运动想象脑电识别系统不能精准提取显著特征、分类识别准确率低等问题,提出一种新的基于自编码器(AE,auto-encoder)降维的Transformer分类识别模型。该方法使用滤...基于运动想象的脑机接口系统一直是海内外研究学者的关注对象。针对传统运动想象脑电识别系统不能精准提取显著特征、分类识别准确率低等问题,提出一种新的基于自编码器(AE,auto-encoder)降维的Transformer分类识别模型。该方法使用滤波器组共空间模式(FBCSP, filter bank common spatial pattern)对数据进行多个频段的特征提取,并利用AE获得降维后的特征矩阵。同时借助Transformer模型的位置编码考虑全局信号特征影响并利用多头自注意力机制考虑特征矩阵的内部关联性,提升系统分类识别效果。与传统的基于线性判别分析(LDA,linear discriminantanalysis)的K-近邻(KNN,K-nearestneighbors)法分类识别系统进行对比,实验表明AE+Transformer模型的分类识别效果优于LDA+KNN系统,说明这种改进后的算法适用于运动想象的二分类。展开更多
文摘Damage detection in structures is performed via vibra-tion based structural identification. Modal information, such as fre-quencies and mode shapes, are widely used for structural dama-ge detection to indicate the health conditions of civil structures.The deep learning algorithm that works on a multiple layer neuralnetwork model termed as deep autoencoder is proposed to learnthe relationship between the modal information and structural stiff-ness parameters. This is achieved via dimension reduction of themodal information feature and a non-linear regression against thestructural stiffness parameters. Numerical tests on a symmetri-cal steel frame model are conducted to generate the data for thetraining and validation, and to demonstrate the efficiency of theproposed approach for vibration based structural damage detec-tion.
文摘目的评估3种深度学习(DL)算法在子宫内膜癌(EC)术后患者高剂量率近距离放射治疗(high-dose-rate brachytherapy,HDR BT)中,自动分割临床靶区(CTV)的应用结果。方法数据集由306名子宫内膜癌术后患者的计算机断层扫描(CT)图像组成,按比例分为训练集(246例)、验证集(30例)和测试集(30例)。比较3种深度卷积神经网络模型(3D U-Net、3D Res U-Net和V-Net)在CTV分割上的性能。采用定量指标分别为戴斯相似性系数(DSC)、豪斯多夫距离(HD)、豪斯多夫距离第95百分位数(HD95%)和交并比(IoU)。结果在测试阶段中,3D U-Net、3D Res U-Net和V-Net分割CTV得到的DSC平均值分别为0.90±0.07、0.95±0.06和0.95±0.06;HD平均值(mm)分别为2.51±1.70、0.96±1.01和0.98±0.95;HD95%平均值(mm)分别为1.33±1.02、0.65±0.91和0.40±0.72,IoU平均值分别为0.85±0.11、0.91±0.09和0.92±0.09。其中,V-Net分割结果与高级临床医生勾画结果更接近,CTV的分割时间<3.2 s,节省了临床医生的工作时间。结论V-Net在CTV分割方面表现最佳,定量指标和临床评估均优于其他模型。该方法与基准真实值高度一致,有效减少医生间差异,缩短治疗时间。
文摘基于运动想象的脑机接口系统一直是海内外研究学者的关注对象。针对传统运动想象脑电识别系统不能精准提取显著特征、分类识别准确率低等问题,提出一种新的基于自编码器(AE,auto-encoder)降维的Transformer分类识别模型。该方法使用滤波器组共空间模式(FBCSP, filter bank common spatial pattern)对数据进行多个频段的特征提取,并利用AE获得降维后的特征矩阵。同时借助Transformer模型的位置编码考虑全局信号特征影响并利用多头自注意力机制考虑特征矩阵的内部关联性,提升系统分类识别效果。与传统的基于线性判别分析(LDA,linear discriminantanalysis)的K-近邻(KNN,K-nearestneighbors)法分类识别系统进行对比,实验表明AE+Transformer模型的分类识别效果优于LDA+KNN系统,说明这种改进后的算法适用于运动想象的二分类。
