[Recognition of motor imagery electroencephalogram based on flicker noise spectroscopy and weighted filter bank common spatial pattern].
Gespeichert in:
| Titel: | [Recognition of motor imagery electroencephalogram based on flicker noise spectroscopy and weighted filter bank common spatial pattern]. |
|---|---|
| Autoren: | Fei K; School of Mechanical Engineering and Rail Transit, Changzhou University, Changzhou, Jiangsu 213164, P.R. China., Cai X; School of Mechanical Engineering and Rail Transit, Changzhou University, Changzhou, Jiangsu 213164, P.R. China., Chen S; School of Mechanical Engineering and Rail Transit, Changzhou University, Changzhou, Jiangsu 213164, P.R. China., Pan L; School of Mechanical Engineering and Rail Transit, Changzhou University, Changzhou, Jiangsu 213164, P.R. China., Wang W; School of Information Science and Engineering, Lanzhou University, Lanzhou 730000, P.R. China. |
| Quelle: | Sheng wu yi xue gong cheng xue za zhi = Journal of biomedical engineering = Shengwu yixue gongchengxue zazhi [Sheng Wu Yi Xue Gong Cheng Xue Za Zhi] 2023 Dec 25; Vol. 40 (6), pp. 1126-1134. |
| Publikationsart: | English Abstract; Journal Article |
| Sprache: | Chinese |
| Info zur Zeitschrift: | Publisher: Sichuan Sheng sheng wu yi xue gong cheng xue hui Country of Publication: China NLM ID: 9426398 Publication Model: Print Cited Medium: Print ISSN: 1001-5515 (Print) Linking ISSN: 10015515 NLM ISO Abbreviation: Sheng Wu Yi Xue Gong Cheng Xue Za Zhi Subsets: MEDLINE |
| Imprint Name(s): | Publication: Chengdu : Sichuan Sheng sheng wu yi xue gong cheng xue hui Original Publication: Chengdu : Sichuan Sheng sheng wu yi xue gong cheng xue hui : Hua xi yi ke da xue : Chengdu ke ji da xue, |
| MeSH-Schlagworte: | Brain-Computer Interfaces*, Imagination ; Signal Processing, Computer-Assisted ; Electroencephalography/methods ; Algorithms ; Spectrum Analysis |
| Abstract: | Due to the high complexity and subject variability of motor imagery electroencephalogram, its decoding is limited by the inadequate accuracy of traditional recognition models. To resolve this problem, a recognition model for motor imagery electroencephalogram based on flicker noise spectrum (FNS) and weighted filter bank common spatial pattern ( w FBCSP) was proposed. First, the FNS method was used to analyze the motor imagery electroencephalogram. Using the second derivative moment as structure function, the ensued precursor time series were generated by using a sliding window strategy, so that hidden dynamic information of transition phase could be captured. Then, based on the characteristic of signal frequency band, the feature of the transition phase precursor time series and reaction phase series were extracted by w FBCSP, generating features representing relevant transition and reaction phase. To make the selected features adapt to subject variability and realize better generalization, algorithm of minimum redundancy maximum relevance was further used to select features. Finally, support vector machine as the classifier was used for the classification. In the motor imagery electroencephalogram recognition, the method proposed in this study yielded an average accuracy of 86.34%, which is higher than the comparison methods. Thus, our proposed method provides a new idea for decoding motor imagery electroencephalogram. |
| References: | IEEE Trans Biomed Eng. 2009 Nov;56(11 Pt 2):2730-3. (PMID: 19605314) Entropy (Basel). 2022 Mar 08;24(3):. (PMID: 35327887) J Neurosci Methods. 2022 Jan 1;365:109378. (PMID: 34626685) J Neurosci Methods. 2022 Apr 01;371:109495. (PMID: 35150764) Sheng Wu Yi Xue Gong Cheng Xue Za Zhi. 2021 Jun 25;38(3):434-446. (PMID: 34180188) Front Neurosci. 2012 Jul 13;6:55. (PMID: 22811657) J Neural Eng. 2010 Apr;7(2):26003. (PMID: 20168003) IEEE Trans Rehabil Eng. 1998 Sep;6(3):316-25. (PMID: 9749909) IEEE Trans Neural Netw Learn Syst. 2021 Feb;32(2):535-545. (PMID: 32745012) Cogn Neurodyn. 2022 Apr;16(2):379-389. (PMID: 35401871) Australas Phys Eng Sci Med. 2015 Mar;38(1):139-49. (PMID: 25649845) J Neural Eng. 2005 Dec;2(4):L14-22. (PMID: 16317224) Neurosci Lett. 1999 Jul 16;269(3):153-6. (PMID: 10454155) Comput Biol Med. 2007 Feb;37(2):183-94. (PMID: 16476421) Front Neurosci. 2012 Mar 29;6:39. (PMID: 22479236) IEEE Trans Biomed Eng. 2007 Jul;54(7):1191-8. (PMID: 17605350) IEEE Trans Pattern Anal Mach Intell. 2005 Aug;27(8):1226-38. (PMID: 16119262) |
| Contributed Indexing: | Keywords: Brain-computer interface; Common spatial pattern; Electroencephalogram signal recognition; Feature selection; Flicker noise spectroscopy Local Abstract: [Publisher, Chinese] 针对运动想象脑电信号复杂度高、受试者个体差异大、传统识别模型精度欠佳的问题，本文提出了基于闪噪谱方法及加权滤波器组共空间模式（ w FBCSP）的运动想象脑电信号识别模型。首先，采用闪噪谱方法对运动想象脑电信号进行解析，以二阶差矩为结构函数，采用滑窗策略生成前兆时间序列，以发掘过渡阶段的隐匿动态变化。其次，从信号频带特点出发，利用 w FBCSP分别对过渡阶段前兆时间序列及反应阶段序列进行特征提取，生成表征过渡阶段及反应阶段的特征向量。进一步，利用最小冗余最大相关算法对特征向量进行局部筛选，使所选特征能自适应于受试者的个体差异，具有更好的泛化性。最后，以支持向量机为分类器进行分类判别。实验结果表明，本文所提方法在运动想象脑电信号识别中取得了86.34%平均分类准确率，较对照方法性能更优，为运动想象脑电信号解码研究提供了新思路。. |
| Entry Date(s): | Date Created: 20231228 Date Completed: 20231229 Latest Revision: 20231230 |
| Update Code: | 20250114 |
| PubMed Central ID: | PMC10753307 |
| DOI: | 10.7507/1001-5515.202302020 |
| PMID: | 38151935 |
| Datenbank: | MEDLINE |
Full Text 
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstract: | Due to the high complexity and subject variability of motor imagery electroencephalogram, its decoding is limited by the inadequate accuracy of traditional recognition models. To resolve this problem, a recognition model for motor imagery electroencephalogram based on flicker noise spectrum (FNS) and weighted filter bank common spatial pattern ( w FBCSP) was proposed. First, the FNS method was used to analyze the motor imagery electroencephalogram. Using the second derivative moment as structure function, the ensued precursor time series were generated by using a sliding window strategy, so that hidden dynamic information of transition phase could be captured. Then, based on the characteristic of signal frequency band, the feature of the transition phase precursor time series and reaction phase series were extracted by w FBCSP, generating features representing relevant transition and reaction phase. To make the selected features adapt to subject variability and realize better generalization, algorithm of minimum redundancy maximum relevance was further used to select features. Finally, support vector machine as the classifier was used for the classification. In the motor imagery electroencephalogram recognition, the method proposed in this study yielded an average accuracy of 86.34%, which is higher than the comparison methods. Thus, our proposed method provides a new idea for decoding motor imagery electroencephalogram. |
|---|---|
| ISSN: | 1001-5515 |
| DOI: | 10.7507/1001-5515.202302020 |