Enhance decoding of lower limb motor imagery‐electroencephalography patterns by Riemannian clustering

Abstract Brain‐Computer Interface (BCI) based on motor imagery (MI) has attracted great interest as a new rehabilitation method for stroke. Riemannian geometry‐based classification algorithms are widely used in MI‐BCI due to their strong robustness and generalization capabilities. However, the clust...

Full description

Saved in:
Bibliographic Details
Main Authors: Xinwei Sun, Tuo Liu, Kun Wang, Lincong Pan, Lin Meng, Xinmin Ding, Weibo Yi, Minpeng Xu, Dong Ming
Format: Article
Language:English
Published: Wiley-VCH 2025-07-01
Series:Interdisciplinary Medicine
Subjects:
Online Access:https://doi.org/10.1002/INMD.20250003
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849728926232870912
author Xinwei Sun
Tuo Liu
Kun Wang
Lincong Pan
Lin Meng
Xinmin Ding
Weibo Yi
Minpeng Xu
Dong Ming
author_facet Xinwei Sun
Tuo Liu
Kun Wang
Lincong Pan
Lin Meng
Xinmin Ding
Weibo Yi
Minpeng Xu
Dong Ming
author_sort Xinwei Sun
collection DOAJ
description Abstract Brain‐Computer Interface (BCI) based on motor imagery (MI) has attracted great interest as a new rehabilitation method for stroke. Riemannian geometry‐based classification algorithms are widely used in MI‐BCI due to their strong robustness and generalization capabilities. However, the clustering performance of current algorithms needs to be improved due to unsuitable clustering criteria for electroencephalography (EEG) characteristics of lower limbs. This study proposed two classification methods based on Riemannian clustering: margin based Riemannian clusters (MBRC) and statistics based Riemannian clusters (SBRC) to address this issue. Our methods divide all samples into subclusters based on the Riemannian distance and innovate clustering criteria. We introduced cluster margin distance and the Riemannian potato algorithm as two clustering criteria to achieve a more robust classification of lower limb MI‐EEG. MBRC and SBRC were tested on an experimental dataset and a public Yi2014 dataset. For the experimental dataset, the average accuracies of MBRC and SBRC were 71.29% and 73.12%, respectively, higher than that of the baseline algorithms. For the Yi2014 dataset, MBRC and SBRC performed better than the comparison algorithms under different training sample numbers, particularly when the number of samples was limited. These findings suggest that the proposed algorithms are more effective for classifying lower limb MI EEGs.
format Article
id doaj-art-8f4dde47e9eb41fa909ba9b9cab60822
institution DOAJ
issn 2832-6245
language English
publishDate 2025-07-01
publisher Wiley-VCH
record_format Article
series Interdisciplinary Medicine
spelling doaj-art-8f4dde47e9eb41fa909ba9b9cab608222025-08-20T03:09:24ZengWiley-VCHInterdisciplinary Medicine2832-62452025-07-0134n/an/a10.1002/INMD.20250003Enhance decoding of lower limb motor imagery‐electroencephalography patterns by Riemannian clusteringXinwei Sun0Tuo Liu1Kun Wang2Lincong Pan3Lin Meng4Xinmin Ding5Weibo Yi6Minpeng Xu7Dong Ming8Tianjin International Joint Research Center for Neural Engineering Academy of Medical Engineering and Translational Medicine Tianjin University Tianjin ChinaSchool of Precision Instruments and Optoelectronics Engineering Tianjin University Tianjin ChinaTianjin International Joint Research Center for Neural Engineering Academy of Medical Engineering and Translational Medicine Tianjin University Tianjin ChinaTianjin International Joint Research Center for Neural Engineering Academy of Medical Engineering and Translational Medicine Tianjin University Tianjin ChinaTianjin International Joint Research Center for Neural Engineering Academy of Medical Engineering and Translational Medicine Tianjin University Tianjin ChinaWest China Tianfu Hospital of Sichuan University Chengdu ChinaBeijing Institute of Mechanical Equipment Beijing ChinaTianjin International Joint Research Center for Neural Engineering Academy of Medical Engineering and Translational Medicine Tianjin University Tianjin ChinaTianjin International Joint Research Center for Neural Engineering Academy of Medical Engineering and Translational Medicine Tianjin University Tianjin ChinaAbstract Brain‐Computer Interface (BCI) based on motor imagery (MI) has attracted great interest as a new rehabilitation method for stroke. Riemannian geometry‐based classification algorithms are widely used in MI‐BCI due to their strong robustness and generalization capabilities. However, the clustering performance of current algorithms needs to be improved due to unsuitable clustering criteria for electroencephalography (EEG) characteristics of lower limbs. This study proposed two classification methods based on Riemannian clustering: margin based Riemannian clusters (MBRC) and statistics based Riemannian clusters (SBRC) to address this issue. Our methods divide all samples into subclusters based on the Riemannian distance and innovate clustering criteria. We introduced cluster margin distance and the Riemannian potato algorithm as two clustering criteria to achieve a more robust classification of lower limb MI‐EEG. MBRC and SBRC were tested on an experimental dataset and a public Yi2014 dataset. For the experimental dataset, the average accuracies of MBRC and SBRC were 71.29% and 73.12%, respectively, higher than that of the baseline algorithms. For the Yi2014 dataset, MBRC and SBRC performed better than the comparison algorithms under different training sample numbers, particularly when the number of samples was limited. These findings suggest that the proposed algorithms are more effective for classifying lower limb MI EEGs.https://doi.org/10.1002/INMD.20250003brain‐computer interface (BCI)motor imagery (MI)Riemannian clusteringsubcluster
spellingShingle Xinwei Sun
Tuo Liu
Kun Wang
Lincong Pan
Lin Meng
Xinmin Ding
Weibo Yi
Minpeng Xu
Dong Ming
Enhance decoding of lower limb motor imagery‐electroencephalography patterns by Riemannian clustering
Interdisciplinary Medicine
brain‐computer interface (BCI)
motor imagery (MI)
Riemannian clustering
subcluster
title Enhance decoding of lower limb motor imagery‐electroencephalography patterns by Riemannian clustering
title_full Enhance decoding of lower limb motor imagery‐electroencephalography patterns by Riemannian clustering
title_fullStr Enhance decoding of lower limb motor imagery‐electroencephalography patterns by Riemannian clustering
title_full_unstemmed Enhance decoding of lower limb motor imagery‐electroencephalography patterns by Riemannian clustering
title_short Enhance decoding of lower limb motor imagery‐electroencephalography patterns by Riemannian clustering
title_sort enhance decoding of lower limb motor imagery electroencephalography patterns by riemannian clustering
topic brain‐computer interface (BCI)
motor imagery (MI)
Riemannian clustering
subcluster
url https://doi.org/10.1002/INMD.20250003
work_keys_str_mv AT xinweisun enhancedecodingoflowerlimbmotorimageryelectroencephalographypatternsbyriemannianclustering
AT tuoliu enhancedecodingoflowerlimbmotorimageryelectroencephalographypatternsbyriemannianclustering
AT kunwang enhancedecodingoflowerlimbmotorimageryelectroencephalographypatternsbyriemannianclustering
AT lincongpan enhancedecodingoflowerlimbmotorimageryelectroencephalographypatternsbyriemannianclustering
AT linmeng enhancedecodingoflowerlimbmotorimageryelectroencephalographypatternsbyriemannianclustering
AT xinminding enhancedecodingoflowerlimbmotorimageryelectroencephalographypatternsbyriemannianclustering
AT weiboyi enhancedecodingoflowerlimbmotorimageryelectroencephalographypatternsbyriemannianclustering
AT minpengxu enhancedecodingoflowerlimbmotorimageryelectroencephalographypatternsbyriemannianclustering
AT dongming enhancedecodingoflowerlimbmotorimageryelectroencephalographypatternsbyriemannianclustering