Results 51 to 60 of about 1,521 (104)

[Diffusion cycle-consistent generative adversarial networks for pelvic active bone marrow segmentation]. [PDF]

Nan Fang Yi Ke Da Xue Xue Bao
Zhuo L, Zeng M, Tan S, Liang T, Xiao W, Zhen X.   +5 more
europepmc   +1 more source

[A motor imagery decoding study integrating differential attention with a multi-scale adaptive temporal convolutional network]. [PDF]

Sheng Wu Yi Xue Gong Cheng Xue Za Zhi
Dong Z, Bao X, Yang Y, Wu J.
europepmc   +1 more source

谐波激励下双层包封空心电抗器的振动噪声特性研究

Gaoya dianqi
广泛应用于高压换流站的干式空心电抗器承受频率含量丰富的谐波电流作用，产生严重的振动及噪声污染问题。干式空心电抗器通常为多层包封结构，文中以一台双层包封的电抗器为例，对其在不同谐波激励下的振动及噪声特性开展了研究，结合磁场力仿真和模态试验分析，明确了内外层包封之间的振动响应的关系，及整体声场的辐射特性。研究结果表明，空心电抗器在不同谐波电流激励下的振动分布受其模态特性主导，内外层包封的振动形态均呈现明显的波峰波谷分布，且与临近固有频率对应的模态振型相似；内外层包封在同一电流激励下的振动幅值和相位相近 ...
高璐, 卢铃, 祝令瑜, 李金宇, 汲胜昌, 李沂新   +5 more
doaj  

[Research progress on quantitative magnetic susceptibility imaging reconstruction method based on improved U-network model]. [PDF]

Sheng Wu Yi Xue Gong Cheng Xue Za Zhi
Yang W, Zhang R, Keung S.
europepmc   +1 more source

[Feature distillation multiple instance learning method based on sequence reorganized Mamba]. [PDF]

Sheng Wu Yi Xue Gong Cheng Xue Za Zhi
Zeng J, Luo W, Zhao J, Huang G, Zhao J, Mai Z, Yan W, Xiao Y, Qin C.   +8 more
europepmc   +1 more source

基于多头注意力机制的CNN-BiLSTM高海拔多因素输电线路可听噪声预测

Gaoya dianqi
为了研究考虑高海拔多环境因素影响下输电线路可听噪声的预测问题，在海拔2 400 m高度点的500 kV同塔双回线路下，搭建了边相外20、30、35 m三处可听噪声观测站，同时利用气象站进行多环境因素指标的数据采集。文中提出了一种基于多头注意力机制（multi-head attention，MHA）的卷积神经网络（convolutional neural network，CNN）—双向长短期记忆网络（bi-directional long short term memory，BiLSTM ...
黄悦华, 张子豪, 陈庆, 刘兴韬, 涂金童   +4 more
doaj  

[Research progress on deep learning-based computer-aided diagnosis of thyroid nodules using ultrasound imaging]. [PDF]

Sheng Wu Yi Xue Gong Cheng Xue Za Zhi
Zhou X, Qiu M, Shang J, Wei G.
europepmc   +1 more source

基于PLO-SVMD与CNN-BiGRU-ATT的直流微电网故障诊断方法

Gaoya dianqi
针对因噪声干扰而导致直流微电网故障识别精度下降的问题，提出一种基于PLO-SVMD与CNN-BiGRU-ATT的故障诊断方法。首先，以最大互信息系数（MIC）为适应度函数，采用极光优化算法（PLO）优化逐次变分模态分解（SVMD）中的惩罚因子，提升信号分解质量与稳定性。其次，依据特征指标筛选出与故障特性高度相关的模态分量，抑制噪声干扰并完成信号重构。最后，构建CNN-BiGRU-ATT分类模型，融合卷积神经网络（CNN）的局部提取能力，双向门控循环单元（BiGRU）的时序建模能力以及注意力机制（ATT ...
毕鑫印, 何山
doaj  

[Full-size diffusion model for adaptive feature medical image fusion]. [PDF]

Sheng Wu Yi Xue Gong Cheng Xue Za Zhi
Di J, Shi S, Wang H, Liang C, Zhu Y.
europepmc   +1 more source

[Clinical value of AI-assisted cardiac auscultation in screening for congenital heart disease in neonates]. [PDF]

Zhejiang Da Xue Xue Bao Yi Xue Ban
Zhang J, Jia W, Zhang Q, Fu C, Xie C, Xu W.   +5 more
europepmc   +1 more source