Combining detrended cross-correlation analysis with Riemannian geometry-based classification for improved brain-computer interface performance
Riemannian geometry-based classification (RGBC) gained popularity in the field of brain-computer interfaces (BCIs) lately, due to its ability to deal with non-stationarities arising in electroencephalography (EEG) data. Domain adaptation, however, is most often performed on sample covariance matrice...
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Frontiers Media S.A.
2024-03-01
|
| Series: | Frontiers in Neuroscience |
| Subjects: | |
| Online Access: | https://www.frontiersin.org/articles/10.3389/fnins.2024.1271831/full |
| _version_ | 1797263074792570880 |
|---|---|
| author | Frigyes Samuel Racz Frigyes Samuel Racz Frigyes Samuel Racz Satyam Kumar Zalan Kaposzta Hussein Alawieh Deland Hu Liu Ruofan Liu Akos Czoch Peter Mukli Peter Mukli Peter Mukli Peter Mukli José del R. Millán José del R. Millán José del R. Millán José del R. Millán |
| author_facet | Frigyes Samuel Racz Frigyes Samuel Racz Frigyes Samuel Racz Satyam Kumar Zalan Kaposzta Hussein Alawieh Deland Hu Liu Ruofan Liu Akos Czoch Peter Mukli Peter Mukli Peter Mukli Peter Mukli José del R. Millán José del R. Millán José del R. Millán José del R. Millán |
| author_sort | Frigyes Samuel Racz |
| collection | DOAJ |
| description | Riemannian geometry-based classification (RGBC) gained popularity in the field of brain-computer interfaces (BCIs) lately, due to its ability to deal with non-stationarities arising in electroencephalography (EEG) data. Domain adaptation, however, is most often performed on sample covariance matrices (SCMs) obtained from EEG data, and thus might not fully account for components affecting covariance estimation itself, such as regional trends. Detrended cross-correlation analysis (DCCA) can be utilized to estimate the covariance structure of such signals, yet it is computationally expensive in its original form. A recently proposed online implementation of DCCA, however, allows for its fast computation and thus makes it possible to employ DCCA in real-time applications. In this study we propose to replace the SCM with the DCCA matrix as input to RGBC and assess its effect on offline and online BCI performance. First we evaluated the proposed decoding pipeline offline on previously recorded EEG data from 18 individuals performing left and right hand motor imagery (MI), and benchmarked it against vanilla RGBC and popular MI-detection approaches. Subsequently, we recruited eight participants (with previous BCI experience) who operated an MI-based BCI (MI-BCI) online using the DCCA-enhanced Riemannian decoder. Finally, we tested the proposed method on a public, multi-class MI-BCI dataset. During offline evaluations the DCCA-based decoder consistently and significantly outperformed the other approaches. Online evaluation confirmed that the DCCA matrix could be computed in real-time even for 22-channel EEG, as well as subjects could control the MI-BCI with high command delivery (normalized Cohen's κ: 0.7409 ± 0.1515) and sample-wise MI detection (normalized Cohen's κ: 0.5200 ± 0.1610). Post-hoc analysis indicated characteristic connectivity patterns under both MI conditions, with stronger connectivity in the hemisphere contralateral to the MI task. Additionally, fractal scaling exponent of neural activity was found increased in the contralateral compared to the ipsilateral motor cortices (C4 and C3 for left and right MI, respectively) in both classes. Combining DCCA with Riemannian geometry-based decoding yields a robust and effective decoder, that not only improves upon the SCM-based approach but can also provide relevant information on the neurophysiological processes behind MI. |
| first_indexed | 2024-04-25T00:07:14Z |
| format | Article |
| id | doaj.art-a075b55974134bd19bca5fcfd8cfe536 |
| institution | Directory Open Access Journal |
| issn | 1662-453X |
| language | English |
| last_indexed | 2024-04-25T00:07:14Z |
| publishDate | 2024-03-01 |
| publisher | Frontiers Media S.A. |
| record_format | Article |
| series | Frontiers in Neuroscience |
| spelling | doaj.art-a075b55974134bd19bca5fcfd8cfe5362024-03-14T04:44:41ZengFrontiers Media S.A.Frontiers in Neuroscience1662-453X2024-03-011810.3389/fnins.2024.12718311271831Combining detrended cross-correlation analysis with Riemannian geometry-based classification for improved brain-computer interface performanceFrigyes Samuel Racz0Frigyes Samuel Racz1Frigyes Samuel Racz2Satyam Kumar3Zalan Kaposzta4Hussein Alawieh5Deland Hu Liu6Ruofan Liu7Akos Czoch8Peter Mukli9Peter Mukli10Peter Mukli11Peter Mukli12José del R. Millán13José del R. Millán14José del R. Millán15José del R. Millán16Department of Neurology, Dell Medical School, The University of Texas at Austin, Austin, TX, United StatesDepartment of Physiology, Faculty of Medicine, Semmelweis University, Budapest, HungaryMulva Clinic for the Neurosciences, Dell Medical School, The University of Texas at Austin, Austin, TX, United StatesChandra Family Department of Electrical and Computer Engineering, Cockrell School of Engineering, The University of Texas at Austin, Austin, TX, United StatesDepartment of Physiology, Faculty of Medicine, Semmelweis University, Budapest, HungaryChandra Family Department of Electrical and Computer Engineering, Cockrell School of Engineering, The University of Texas at Austin, Austin, TX, United StatesChandra Family Department of Electrical and Computer Engineering, Cockrell School of Engineering, The University of Texas at Austin, Austin, TX, United StatesChandra Family Department of Electrical and Computer Engineering, Cockrell School of Engineering, The University of Texas at Austin, Austin, TX, United StatesDepartment of Physiology, Faculty of Medicine, Semmelweis University, Budapest, HungaryDepartment of Physiology, Faculty of Medicine, Semmelweis University, Budapest, HungaryOklahoma Center for Geroscience and Healthy Brain Aging, University of Oklahoma Health Sciences Center, Oklahoma City, OK, United StatesVascular Cognitive Impairment and Neurodegeneration Program, Department of Neurosurgery, University of Oklahoma Health Sciences Center, Oklahoma City, OK, United StatesInternational Training Program in Geroscience, Doctoral School of Basic and Translational Medicine/Department of Public Health, Semmelweis University, Budapest, HungaryDepartment of Neurology, Dell Medical School, The University of Texas at Austin, Austin, TX, United StatesMulva Clinic for the Neurosciences, Dell Medical School, The University of Texas at Austin, Austin, TX, United StatesChandra Family Department of Electrical and Computer Engineering, Cockrell School of Engineering, The University of Texas at Austin, Austin, TX, United StatesDepartment of Biomedical Engineering, Cockrell School of Engineering, The University of Texas at Austin, Austin, TX, United StatesRiemannian geometry-based classification (RGBC) gained popularity in the field of brain-computer interfaces (BCIs) lately, due to its ability to deal with non-stationarities arising in electroencephalography (EEG) data. Domain adaptation, however, is most often performed on sample covariance matrices (SCMs) obtained from EEG data, and thus might not fully account for components affecting covariance estimation itself, such as regional trends. Detrended cross-correlation analysis (DCCA) can be utilized to estimate the covariance structure of such signals, yet it is computationally expensive in its original form. A recently proposed online implementation of DCCA, however, allows for its fast computation and thus makes it possible to employ DCCA in real-time applications. In this study we propose to replace the SCM with the DCCA matrix as input to RGBC and assess its effect on offline and online BCI performance. First we evaluated the proposed decoding pipeline offline on previously recorded EEG data from 18 individuals performing left and right hand motor imagery (MI), and benchmarked it against vanilla RGBC and popular MI-detection approaches. Subsequently, we recruited eight participants (with previous BCI experience) who operated an MI-based BCI (MI-BCI) online using the DCCA-enhanced Riemannian decoder. Finally, we tested the proposed method on a public, multi-class MI-BCI dataset. During offline evaluations the DCCA-based decoder consistently and significantly outperformed the other approaches. Online evaluation confirmed that the DCCA matrix could be computed in real-time even for 22-channel EEG, as well as subjects could control the MI-BCI with high command delivery (normalized Cohen's κ: 0.7409 ± 0.1515) and sample-wise MI detection (normalized Cohen's κ: 0.5200 ± 0.1610). Post-hoc analysis indicated characteristic connectivity patterns under both MI conditions, with stronger connectivity in the hemisphere contralateral to the MI task. Additionally, fractal scaling exponent of neural activity was found increased in the contralateral compared to the ipsilateral motor cortices (C4 and C3 for left and right MI, respectively) in both classes. Combining DCCA with Riemannian geometry-based decoding yields a robust and effective decoder, that not only improves upon the SCM-based approach but can also provide relevant information on the neurophysiological processes behind MI.https://www.frontiersin.org/articles/10.3389/fnins.2024.1271831/fullbrain-computer interfacedetrended cross-correlation analysisReimannian geometrymotor imagerydetrended fluctuation analysisfractal connectivity |
| spellingShingle | Frigyes Samuel Racz Frigyes Samuel Racz Frigyes Samuel Racz Satyam Kumar Zalan Kaposzta Hussein Alawieh Deland Hu Liu Ruofan Liu Akos Czoch Peter Mukli Peter Mukli Peter Mukli Peter Mukli José del R. Millán José del R. Millán José del R. Millán José del R. Millán Combining detrended cross-correlation analysis with Riemannian geometry-based classification for improved brain-computer interface performance Frontiers in Neuroscience brain-computer interface detrended cross-correlation analysis Reimannian geometry motor imagery detrended fluctuation analysis fractal connectivity |
| title | Combining detrended cross-correlation analysis with Riemannian geometry-based classification for improved brain-computer interface performance |
| title_full | Combining detrended cross-correlation analysis with Riemannian geometry-based classification for improved brain-computer interface performance |
| title_fullStr | Combining detrended cross-correlation analysis with Riemannian geometry-based classification for improved brain-computer interface performance |
| title_full_unstemmed | Combining detrended cross-correlation analysis with Riemannian geometry-based classification for improved brain-computer interface performance |
| title_short | Combining detrended cross-correlation analysis with Riemannian geometry-based classification for improved brain-computer interface performance |
| title_sort | combining detrended cross correlation analysis with riemannian geometry based classification for improved brain computer interface performance |
| topic | brain-computer interface detrended cross-correlation analysis Reimannian geometry motor imagery detrended fluctuation analysis fractal connectivity |
| url | https://www.frontiersin.org/articles/10.3389/fnins.2024.1271831/full |
| work_keys_str_mv | AT frigyessamuelracz combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT frigyessamuelracz combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT frigyessamuelracz combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT satyamkumar combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT zalankaposzta combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT husseinalawieh combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT delandhuliu combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT ruofanliu combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT akosczoch combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT petermukli combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT petermukli combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT petermukli combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT petermukli combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT josedelrmillan combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT josedelrmillan combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT josedelrmillan combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance AT josedelrmillan combiningdetrendedcrosscorrelationanalysiswithriemanniangeometrybasedclassificationforimprovedbraincomputerinterfaceperformance |