Neural correlates of user learning during long-term BCI training for the Cybathlon competition.
Brain-computer interface
Cybathlon
Long-term evaluation
Motor imagery
Mutual learning
Riemann geometry
User learning
Journal
Journal of neuroengineering and rehabilitation
ISSN: 1743-0003
Titre abrégé: J Neuroeng Rehabil
Pays: England
ID NLM: 101232233
Informations de publication
Date de publication:
05 07 2022
05 07 2022
Historique:
received:
30
09
2021
accepted:
22
06
2022
entrez:
5
7
2022
pubmed:
6
7
2022
medline:
8
7
2022
Statut:
epublish
Résumé
Brain-computer interfaces (BCIs) are systems capable of translating human brain patterns, measured through electroencephalography (EEG), into commands for an external device. Despite the great advances in machine learning solutions to enhance the performance of BCI decoders, the translational impact of this technology remains elusive. The reliability of BCIs is often unsatisfactory for end-users, limiting their application outside a laboratory environment. We present the analysis on the data acquired from an end-user during the preparation for two Cybathlon competitions, where our pilot won the gold medal twice in a row. These data are of particular interest given the mutual learning approach adopted during the longitudinal training phase (8 months), the long training break in between the two events (1 year) and the demanding evaluation scenario. A multifaceted perspective on long-term user learning is proposed: we enriched the information gathered through conventional metrics (e.g., accuracy, application performances) by investigating novel neural correlates of learning in different neural domains. First, we showed that by focusing the training on user learning, the pilot was capable of significantly improving his performance over time even with infrequent decoder re-calibrations. Second, we revealed that the analysis of the within-class modifications of the pilot's neural patterns in the Riemannian domain is more effective in tracking the acquisition and the stabilization of BCI skills, especially after the 1-year break. These results further confirmed the key role of mutual learning in the acquisition of BCI skills, and particularly highlighted the importance of user learning as a key to enhance BCI reliability. We firmly believe that our work may open new perspectives and fuel discussions in the BCI field to shift the focus of future research: not only to the machine learning of the decoder, but also in investigating novel training procedures to boost the user learning and the stability of the BCI skills in the long-term. To this end, the analyses and the metrics proposed could be used to monitor the user learning during training and provide a marker guiding the decoder re-calibration to maximize the mutual adaptation of the user to the BCI system.
Sections du résumé
BACKGROUND
Brain-computer interfaces (BCIs) are systems capable of translating human brain patterns, measured through electroencephalography (EEG), into commands for an external device. Despite the great advances in machine learning solutions to enhance the performance of BCI decoders, the translational impact of this technology remains elusive. The reliability of BCIs is often unsatisfactory for end-users, limiting their application outside a laboratory environment.
METHODS
We present the analysis on the data acquired from an end-user during the preparation for two Cybathlon competitions, where our pilot won the gold medal twice in a row. These data are of particular interest given the mutual learning approach adopted during the longitudinal training phase (8 months), the long training break in between the two events (1 year) and the demanding evaluation scenario. A multifaceted perspective on long-term user learning is proposed: we enriched the information gathered through conventional metrics (e.g., accuracy, application performances) by investigating novel neural correlates of learning in different neural domains.
RESULTS
First, we showed that by focusing the training on user learning, the pilot was capable of significantly improving his performance over time even with infrequent decoder re-calibrations. Second, we revealed that the analysis of the within-class modifications of the pilot's neural patterns in the Riemannian domain is more effective in tracking the acquisition and the stabilization of BCI skills, especially after the 1-year break. These results further confirmed the key role of mutual learning in the acquisition of BCI skills, and particularly highlighted the importance of user learning as a key to enhance BCI reliability.
CONCLUSION
We firmly believe that our work may open new perspectives and fuel discussions in the BCI field to shift the focus of future research: not only to the machine learning of the decoder, but also in investigating novel training procedures to boost the user learning and the stability of the BCI skills in the long-term. To this end, the analyses and the metrics proposed could be used to monitor the user learning during training and provide a marker guiding the decoder re-calibration to maximize the mutual adaptation of the user to the BCI system.
Identifiants
pubmed: 35790978
doi: 10.1186/s12984-022-01047-x
pii: 10.1186/s12984-022-01047-x
pmc: PMC9254548
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
69Informations de copyright
© 2022. The Author(s).
Références
Clin Neurophysiol. 2007 Mar;118(3):480-94
pubmed: 17169606
Front Hum Neurosci. 2021 Mar 18;15:635653
pubmed: 33815081
Sci Data. 2021 Apr 1;8(1):98
pubmed: 33795705
Brain Topogr. 2020 Nov;33(6):710-719
pubmed: 32685998
J Neurosci. 2016 Mar 23;36(12):3623-32
pubmed: 27013690
PLoS Biol. 2018 May 10;16(5):e2003787
pubmed: 29746465
J Neural Eng. 2016 Jun;13(3):036018
pubmed: 27152498
Neuroimage. 2010 Jul 15;51(4):1303-9
pubmed: 20303409
Brain Topogr. 2010 Jun;23(2):194-8
pubmed: 19946737
J Neuroeng Rehabil. 2020 Apr 25;17(1):57
pubmed: 32334608
J Neural Eng. 2016 Jun;13(3):036024
pubmed: 27172246
Rev Sci Instrum. 2016 Aug;87(8):085110
pubmed: 27587163
Front Integr Neurosci. 2015 Jun 10;9:40
pubmed: 26113812
Hum Mov Sci. 2015 Oct;43:107-17
pubmed: 26253223
Ann Clin Transl Neurol. 2018 Mar 25;5(5):651-663
pubmed: 29761128
Neural Comput. 2013 Jul;25(7):1693-731
pubmed: 23607558
J Neural Eng. 2013 Oct;10(5):056005
pubmed: 23918061
Proc Natl Acad Sci U S A. 2008 Dec 9;105(49):19486-91
pubmed: 19047633
Front Hum Neurosci. 2021 Mar 25;15:647908
pubmed: 33841120
J Neural Eng. 2011 Apr;8(2):025009
pubmed: 21436515
IEEE Trans Neural Syst Rehabil Eng. 2012 May;20(3):313-9
pubmed: 22481835
Neurosci Lett. 2009 Oct 2;462(1):94-8
pubmed: 19545601
Psychon Bull Rev. 2013 Feb;20(1):21-53
pubmed: 23132605
Sci Robot. 2019 Jun 26;4(31):
pubmed: 31656937
PLoS Biol. 2003 Nov;1(2):E42
pubmed: 14624244
Nat Rev Neurol. 2016 Sep;12(9):513-25
pubmed: 27539560
Front Hum Neurosci. 2013 Sep 17;7:568
pubmed: 24062669
Front Neuroeng. 2010 Mar 30;3:3
pubmed: 20407639
IEEE Trans Neural Syst Rehabil Eng. 2003 Jun;11(2):145-7
pubmed: 12899258
Front Neurosci. 2010 Jun 18;4:
pubmed: 20859445
Front Neurosci. 2012 Nov 19;6:169
pubmed: 23181009
IEEE Trans Neural Syst Rehabil Eng. 2020 Nov;28(11):2525-2536
pubmed: 32997632
Front Hum Neurosci. 2019 Apr 17;13:128
pubmed: 31057380
Clin Neurophysiol. 2002 Jun;113(6):767-91
pubmed: 12048038
IEEE Trans Biomed Eng. 2018 May;65(5):1107-1116
pubmed: 28841546
Front Hum Neurosci. 2021 Jun 15;15:648275
pubmed: 34211380
Nat Commun. 2018 Jun 20;9(1):2421
pubmed: 29925890
Front Comput Neurosci. 2013 Nov 05;7:157
pubmed: 24204342
Nature. 1999 Mar 25;398(6725):297-8
pubmed: 10192330
Front Hum Neurosci. 2021 Feb 26;15:635777
pubmed: 33716698
Neural Comput. 2011 Mar;23(3):791-816
pubmed: 21162666
Front Neurosci. 2018 Sep 11;12:624
pubmed: 30271318
Brain Comput Interfaces (Abingdon). 2017;4(1-2):60-73
pubmed: 29629393
PLoS One. 2011 Jan 31;6(1):e14634
pubmed: 21297944
PLoS One. 2014 Jul 28;9(7):e102504
pubmed: 25068464
Annu Int Conf IEEE Eng Med Biol Soc. 2009;2009:3353-6
pubmed: 19964077
Neurophysiol Clin. 2000 Feb;30(1):5-19
pubmed: 10740792
IEEE Trans Neural Syst Rehabil Eng. 2010 Apr;18(2):107-16
pubmed: 20083463
Proc Natl Acad Sci U S A. 2013 Jun 25;110(26):10818-23
pubmed: 23754426
IEEE Trans Biomed Eng. 2006 Jun;53(6):1214-9
pubmed: 16761852
Neuron. 2014 Jun 18;82(6):1380-93
pubmed: 24945777
PLoS Biol. 2009 Jul;7(7):e1000153
pubmed: 19621062