{"title":"Transformed wavelets for motor imagery EEG classification using hybrid CNN-modified vision transformer: an exploratory study of MI EEG.","authors":"Balendra, Neeraj Sharma, Shiru Sharma","doi":"10.1080/10255842.2025.2563351","DOIUrl":null,"url":null,"abstract":"<p><p>Wavelets capture signal characteristics across time and frequency, but traditional wavelets suffer from high time-bandwidth products (TBP), limiting feature discrimination in EEG classification. We propose transformed wavelets with improved TBP and frequency bandwidth, outperforming Morlet by 0.04 and 0.20, respectively. Using datasets BCI Competition IV 2a, 2b, and CLA, we evaluated both fundamental and transformed wavelets with a modified vision transformer (MViT). Enhanced scalograms generated through local mean and principal component analysis (PCA) consistently outperformed raw scalograms. A hybrid convolutional neural network (CNN)-MViT achieved 82.35% inter-subject and 89.02% intra-subject accuracy, with 3-4% average gains in motor imagery EEG decoding.</p>","PeriodicalId":50640,"journal":{"name":"Computer Methods in Biomechanics and Biomedical Engineering","volume":" ","pages":"1-18"},"PeriodicalIF":1.6000,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computer Methods in Biomechanics and Biomedical Engineering","FirstCategoryId":"5","ListUrlMain":"https://doi.org/10.1080/10255842.2025.2563351","RegionNum":4,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}
引用次数: 0
Abstract
Wavelets capture signal characteristics across time and frequency, but traditional wavelets suffer from high time-bandwidth products (TBP), limiting feature discrimination in EEG classification. We propose transformed wavelets with improved TBP and frequency bandwidth, outperforming Morlet by 0.04 and 0.20, respectively. Using datasets BCI Competition IV 2a, 2b, and CLA, we evaluated both fundamental and transformed wavelets with a modified vision transformer (MViT). Enhanced scalograms generated through local mean and principal component analysis (PCA) consistently outperformed raw scalograms. A hybrid convolutional neural network (CNN)-MViT achieved 82.35% inter-subject and 89.02% intra-subject accuracy, with 3-4% average gains in motor imagery EEG decoding.
小波可以捕获跨时间和频率的信号特征，但传统的小波具有高时间带宽积（TBP），限制了脑电信号分类中的特征识别。我们提出了改进TBP和频率带宽的变换小波，分别比Morlet高0.04和0.20。使用BCI Competition IV 2a， 2b和CLA数据集，我们使用改进的视觉转换器（MViT）评估基本小波和变换小波。通过局部均值和主成分分析（PCA）生成的增强尺度图始终优于原始尺度图。混合卷积神经网络(CNN)-MViT在被试间和被试内的解码准确率分别达到82.35%和89.02%，运动意象EEG解码平均提高3-4%。
期刊介绍：
The primary aims of Computer Methods in Biomechanics and Biomedical Engineering are to provide a means of communicating the advances being made in the areas of biomechanics and biomedical engineering and to stimulate interest in the continually emerging computer based technologies which are being applied in these multidisciplinary subjects. Computer Methods in Biomechanics and Biomedical Engineering will also provide a focus for the importance of integrating the disciplines of engineering with medical technology and clinical expertise. Such integration will have a major impact on health care in the future.
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号
