一种基于Transformer脑区时序分析的DRDS脑电信号识别方法

本发明公开了一种基于Transformer脑区时序分析的DRDS脑电信号识别方法，涉及脑电信号识别技术领域。本发明包括如下步骤：对受试者的EEG数据进行预处理，将预处理后的EEG信号进行切片，作为网络的输入样本；使用时间卷积和维度变换提取EEG信号的特征，将提取的特征送入包含Transformer Encoder结构的脑区Transformer模块提取空间特征；将空间特征进行转置，送入包含Transformer Encoder结构的时序Transformer模块进行全局自注意力特征提取，提取时序特征；构建时空多尺度卷积融合模块，得到高级EEG时空特征，完成EEG信号的分类。本发明有效地利用了脑...

Full description

Saved in:
Bibliographic Details
Format Patent
LanguageChinese
Published 17.05.2024
Subjects
DIAGNOSIS
HUMAN NECESSITIES
HYGIENE
IDENTIFICATION
MEDICAL OR VETERINARY SCIENCE
SURGERY
Online AccessGet full text

Cover

More Information
Summary:本发明公开了一种基于Transformer脑区时序分析的DRDS脑电信号识别方法，涉及脑电信号识别技术领域。本发明包括如下步骤：对受试者的EEG数据进行预处理，将预处理后的EEG信号进行切片，作为网络的输入样本；使用时间卷积和维度变换提取EEG信号的特征，将提取的特征送入包含Transformer Encoder结构的脑区Transformer模块提取空间特征；将空间特征进行转置，送入包含Transformer Encoder结构的时序Transformer模块进行全局自注意力特征提取，提取时序特征；构建时空多尺度卷积融合模块，得到高级EEG时空特征，完成EEG信号的分类。本发明有效地利用了脑电信号电极之间的空间联系，充分挖掘上下文之间的时序信息，提高了分类准确率。 The invention discloses a DRDS electroencephalogram signal recognition method based on Transform brain region time sequence analysis, and relates to the technical field of electroencephalogram signal recognition. The method comprises the following steps: preprocessing EEG data of a subject, and slicing the preprocessed EEG signal as an input sample of a network; the method comprises the following steps: extracting features of an EEG signal by using time convolution and dimension transformation, and sending the extracted features into a brain region Transform module containing a Transform Encoder structure to extract spatial features; the spatial features are transposed, the tra
Bibliography:Application Number: CN202211588317