[Effect of AI-assisted compressed sensing acceleration on MRI radiomic feature extraction and staging model performance for nasopharyngeal carcinoma]. [PDF]
Li X, Xu G, Liu J, Feng Y.
europepmc +1 more source
[Sampling intervals dependent feature extraction for state transfer networks of epileptic signals]. [PDF]
Zhang L, Yan S, Gu C.
europepmc +1 more source
[Fatigue feature extraction and classification algorithm of forehead single-channel electroencephalography signals]. [PDF]
Yang H, Liu Y, Xia L.
europepmc +1 more source
针对电力设备缺陷文本信息的知识挖掘与分析任务中存在缺陷文本特征信息提取不足、缺陷文本分类精度不够的问题,提出一种基于BERT(bidirectional encoder representations from transformers)的双分支特征融合的电力设备缺陷文本分类模型。首先,对缺陷文本数据进行预处理,删除异常缺陷文本,并归纳了电力设备缺陷文本特点;然后,采用BERT模型作为文本编码器,将文本转化为向量后分别输入至BiLSTM-Attention(attention-based ...
张中文 +4 more
doaj
[Heart sound classification algorithm based on bispectral feature extraction and convolutional neural networks]. [PDF]
Peng L, Quan H.
europepmc +1 more source
为了提高电力系统暂态电压在系统拓扑结构发生变化时能够稳定评估,以及提高在时空方面的特征提取能力,提出一种图卷积网络与循环神经网络相融合的方法。首先,引入图卷积网络对电力数据进行图表示,将电力系统建模为网络结构,自动学习电压节点的特征表示。其次,提出使用循环神经网络来处理暂态电压数据的时间依赖关系,捕捉暂态电压数据的时序特征。然后,提出自适应增强模块,用于将两个输出特征表示相互融合,提高模型在系统拓扑结构上的时空特征提取能力。最后,通过算例验证表明,相比于传统的评估模型,所提方法具有更高的预测精度和有效性。
徐焕 +5 more
doaj
[Multi-modal physiological time-frequency feature extraction network for accurate sleep stage classification]. [PDF]
Hu K, Chen J, Zhang P, Xue W, Xie J.
europepmc +1 more source
[Fully Automatic Glioma Segmentation Algorithm of Magnetic Resonance Imaging Based on 3D-UNet With More Global Contextual Feature Extraction: An Improvement on Insufficient Extraction of Global Features]. [PDF]
Tian H, Wang Y, Ji Y, Rahman MM.
europepmc +1 more source
[Liver lesion classification in multi-modal magnetic resonance imaging via medical foundation segmentation models and radiomics integration]. [PDF]
Liu Y +5 more
europepmc +1 more source
[Dual-stream temporal-frequency convolutional network with additive attention for auditory attention decoding]. [PDF]
Lin X, Zhang S, Huang S.
europepmc +1 more source

