Classifier-based latency estimation: a novel way to estimate and predict BCI accuracy

Brain-computer interfaces (BCIs) that detect event-related potentials (ERPs) rely on classification schemes that are vulnerable to latency jitter, a phenomenon known to occur with ERPs such as the P300 response. The objective of this work was to investigate the role that latency jitter plays in BCI...

Full description

Saved in:
Bibliographic Details
Published in:Journal of neural engineering Vol. 10; no. 1; p. 016006
Main Authors: Thompson, David E, Warschausky, Seth, Huggins, Jane E
Format: Journal Article
Language:English
Published: England 01.02.2013
Subjects:
Adolescent
Adult
Aged
Brain - physiology
Brain-Computer Interfaces - classification
Brain-Computer Interfaces - standards
Female
Humans
Male
Middle Aged
Predictive Value of Tests
Reaction Time - physiology
Young Adult
ISSN:1741-2552, 1741-2552
Online Access:Get more information
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract Brain-computer interfaces (BCIs) that detect event-related potentials (ERPs) rely on classification schemes that are vulnerable to latency jitter, a phenomenon known to occur with ERPs such as the P300 response. The objective of this work was to investigate the role that latency jitter plays in BCI classification. We developed a novel method, classifier-based latency estimation (CBLE), based on a generalization of Woody filtering. The technique works by presenting the time-shifted data to the classifier, and using the time shift that corresponds to the maximal classifier score. The variance of CBLE estimates correlates significantly (p < 10(-42)) with BCI accuracy in the Farwell-Donchin BCI paradigm. Additionally, CBLE predicts same-day accuracy, even from small datasets or datasets that have already been used for classifier training, better than the accuracy on the small dataset (p < 0.05). The technique should be relatively classifier-independent, and the results were confirmed on two linear classifiers. The results suggest that latency jitter may be an important cause of poor BCI performance, and methods that correct for latency jitter may improve that performance. CBLE can also be used to decrease the amount of data needed for accuracy estimation, allowing research on effects with shorter timescales.
AbstractList Brain-computer interfaces (BCIs) that detect event-related potentials (ERPs) rely on classification schemes that are vulnerable to latency jitter, a phenomenon known to occur with ERPs such as the P300 response. The objective of this work was to investigate the role that latency jitter plays in BCI classification.OBJECTIVEBrain-computer interfaces (BCIs) that detect event-related potentials (ERPs) rely on classification schemes that are vulnerable to latency jitter, a phenomenon known to occur with ERPs such as the P300 response. The objective of this work was to investigate the role that latency jitter plays in BCI classification.We developed a novel method, classifier-based latency estimation (CBLE), based on a generalization of Woody filtering. The technique works by presenting the time-shifted data to the classifier, and using the time shift that corresponds to the maximal classifier score.APPROACHWe developed a novel method, classifier-based latency estimation (CBLE), based on a generalization of Woody filtering. The technique works by presenting the time-shifted data to the classifier, and using the time shift that corresponds to the maximal classifier score.The variance of CBLE estimates correlates significantly (p < 10(-42)) with BCI accuracy in the Farwell-Donchin BCI paradigm. Additionally, CBLE predicts same-day accuracy, even from small datasets or datasets that have already been used for classifier training, better than the accuracy on the small dataset (p < 0.05). The technique should be relatively classifier-independent, and the results were confirmed on two linear classifiers.MAIN RESULTSThe variance of CBLE estimates correlates significantly (p < 10(-42)) with BCI accuracy in the Farwell-Donchin BCI paradigm. Additionally, CBLE predicts same-day accuracy, even from small datasets or datasets that have already been used for classifier training, better than the accuracy on the small dataset (p < 0.05). The technique should be relatively classifier-independent, and the results were confirmed on two linear classifiers.The results suggest that latency jitter may be an important cause of poor BCI performance, and methods that correct for latency jitter may improve that performance. CBLE can also be used to decrease the amount of data needed for accuracy estimation, allowing research on effects with shorter timescales.SIGNIFICANCEThe results suggest that latency jitter may be an important cause of poor BCI performance, and methods that correct for latency jitter may improve that performance. CBLE can also be used to decrease the amount of data needed for accuracy estimation, allowing research on effects with shorter timescales.
Brain-computer interfaces (BCIs) that detect event-related potentials (ERPs) rely on classification schemes that are vulnerable to latency jitter, a phenomenon known to occur with ERPs such as the P300 response. The objective of this work was to investigate the role that latency jitter plays in BCI classification. We developed a novel method, classifier-based latency estimation (CBLE), based on a generalization of Woody filtering. The technique works by presenting the time-shifted data to the classifier, and using the time shift that corresponds to the maximal classifier score. The variance of CBLE estimates correlates significantly (p < 10(-42)) with BCI accuracy in the Farwell-Donchin BCI paradigm. Additionally, CBLE predicts same-day accuracy, even from small datasets or datasets that have already been used for classifier training, better than the accuracy on the small dataset (p < 0.05). The technique should be relatively classifier-independent, and the results were confirmed on two linear classifiers. The results suggest that latency jitter may be an important cause of poor BCI performance, and methods that correct for latency jitter may improve that performance. CBLE can also be used to decrease the amount of data needed for accuracy estimation, allowing research on effects with shorter timescales.
Author Warschausky, Seth
Huggins, Jane E
Thompson, David E
Author_xml – sequence: 1
  givenname: David E
  surname: Thompson
  fullname: Thompson, David E
  email: dthomp@umich.edu
  organization: Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI, USA. dthomp@umich.edu
– sequence: 2
  givenname: Seth
  surname: Warschausky
  fullname: Warschausky, Seth
– sequence: 3
  givenname: Jane E
  surname: Huggins
  fullname: Huggins, Jane E
BackLink https://www.ncbi.nlm.nih.gov/pubmed/23234797$$D View this record in MEDLINE/PubMed
BookMark eNpNUMtqwzAQFCWlebSf0KJjL270sCW7t9b0EQj00pzNSlqBiyOnlt3iv4-hCRQWdpgZhp1dklloAxJyy9kDZ3m-5jrlicgUW_Np1owrxtQFWZz4TMz-4TlZxvjFmOS6YFdkLqSQqS70guzKBmKsfY1dYiCiow30GOxIMfb1Hvq6DY8UaGh_sKG_MNK-PUtIITh66NDVtqfP5YaCtUMHdrwmlx6aiDenvSK715fP8j3ZfrxtyqdtYtNc9YkRmDMupHJOm4JLAOs0ypR5o4CDLIRzrii8Ql2ozEtvMiaQe2Mty_LciRW5_8s9dO33MJ1V7etosWkgYDvEik-F8yxLhZ6sdyfrYPboqkM3VejG6vwKcQR7J2NJ
CitedBy_id crossref_primary_10_1088_1741_2560_11_3_035008
crossref_primary_10_3390_brainsci10100734
crossref_primary_10_1109_JBHI_2018_2883458
crossref_primary_10_1080_2326263X_2020_1734401
crossref_primary_10_1088_1741_2552_ad5ec0
crossref_primary_10_1080_01621459_2025_2498088
crossref_primary_10_1080_2326263X_2017_1338010
crossref_primary_10_1080_2326263X_2021_2014678
crossref_primary_10_3389_fnhum_2022_930433
crossref_primary_10_3390_signals5010002
crossref_primary_10_1016_j_compbiomed_2023_107658
crossref_primary_10_1109_ACCESS_2020_3000187
crossref_primary_10_1016_j_neuroimage_2019_04_049
crossref_primary_10_1088_1741_2552_aa8416
crossref_primary_10_1080_2326263X_2016_1203629
crossref_primary_10_1088_1741_2560_13_6_066007
crossref_primary_10_1016_j_neuroimage_2016_01_019
crossref_primary_10_3389_fnhum_2018_00165
crossref_primary_10_1016_j_neucom_2014_07_043
ContentType Journal Article
DBID CGR
CUY
CVF
ECM
EIF
NPM
7X8
DOI 10.1088/1741-2560/10/1/016006
DatabaseName Medline
MEDLINE
MEDLINE (Ovid)
MEDLINE
MEDLINE
PubMed
MEDLINE - Academic
DatabaseTitle MEDLINE
Medline Complete
MEDLINE with Full Text
PubMed
MEDLINE (Ovid)
MEDLINE - Academic
DatabaseTitleList MEDLINE - Academic
MEDLINE
Database_xml – sequence: 1
  dbid: NPM
  name: PubMed
  url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed
  sourceTypes: Index Database
– sequence: 2
  dbid: 7X8
  name: MEDLINE - Academic
  url: https://search.proquest.com/medline
  sourceTypes: Aggregation Database
DeliveryMethod no_fulltext_linktorsrc
Discipline Anatomy & Physiology
EISSN 1741-2552
ExternalDocumentID 23234797
Genre Research Support, U.S. Gov't, Non-P.H.S
Journal Article
Research Support, N.I.H., Extramural
GrantInformation_xml – fundername: NICHD NIH HHS
  grantid: R21 HD054697
– fundername: NICHD NIH HHS
  grantid: R21HD054697
GroupedDBID ---
02O
1JI
1WK
4.4
53G
5B3
5GY
5VS
5ZH
7.M
7.Q
AAGCD
AAJIO
AAJKP
AALHV
AATNI
ABHWH
ABJNI
ABQJV
ABVAM
ACAFW
ACARI
ACGFS
ACHIP
ADEQX
AEFHF
AENEX
AERVB
AFYNE
AGQPQ
AHSEE
AKPSB
ALMA_UNASSIGNED_HOLDINGS
AOAED
ARNYC
ASPBG
ATQHT
AVWKF
AZFZN
BBWZM
CEBXE
CGR
CJUJL
CRLBU
CS3
CUY
CVF
DU5
EBS
ECM
EDWGO
EIF
EJD
EMSAF
EPQRW
EQZZN
F5P
FEDTE
HVGLF
IHE
IJHAN
IOP
IZVLO
JCGBZ
KOT
LAP
M45
N5L
N9A
NPM
NT-
NT.
P2P
PJBAE
Q02
RIN
RNS
RO9
ROL
RPA
S3P
SY9
W28
XPP
7X8
AEINN
ID FETCH-LOGICAL-c486t-b2e801236dd7b913aacd7e340fb6a1a392ddd99f6e7965f3fb502e1fbcc0588d2
IEDL.DBID 7X8
ISICitedReferencesCount 32
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000316245500008&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 1741-2552
IngestDate Thu Oct 02 06:34:18 EDT 2025
Mon Jul 21 05:38:39 EDT 2025
IsDoiOpenAccess false
IsOpenAccess true
IsPeerReviewed true
IsScholarly true
Issue 1
Language English
LinkModel DirectLink
MergedId FETCHMERGED-LOGICAL-c486t-b2e801236dd7b913aacd7e340fb6a1a392ddd99f6e7965f3fb502e1fbcc0588d2
Notes ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
OpenAccessLink https://www.ncbi.nlm.nih.gov/pmc/articles/3650625
PMID 23234797
PQID 1317855427
PQPubID 23479
ParticipantIDs proquest_miscellaneous_1317855427
pubmed_primary_23234797
PublicationCentury 2000
PublicationDate 2013-02-01
PublicationDateYYYYMMDD 2013-02-01
PublicationDate_xml – month: 02
  year: 2013
  text: 2013-02-01
  day: 01
PublicationDecade 2010
PublicationPlace England
PublicationPlace_xml – name: England
PublicationTitle Journal of neural engineering
PublicationTitleAlternate J Neural Eng
PublicationYear 2013
References 19041343 - J Neurosci Methods. 2009 Mar 15;177(2):488-96
4695775 - Exp Brain Res. 1973 Mar 19;16(5):455-65
10731765 - Psychophysiology. 2000 Mar;37(2):127-52
20405196 - Brain Topogr. 2010 Jun;23(2):180-5
1464675 - J Clin Neurophysiol. 1992 Oct;9(4):456-79
19545601 - Neurosci Lett. 2009 Oct 2;462(1):94-8
17573239 - Clin Neurophysiol. 2007 Oct;118(10):2128-48
22143614 - J Neurol. 2012 Jun;259(6):1191-8
20509913 - Behav Brain Funct. 2010;6:28
20214929 - Neurosci Res. 2010 Jun;67(2):172-80
22030141 - Clin Neurophysiol. 2012 Jun;123(6):1123-30
12048038 - Clin Neurophysiol. 2002 Jun;113(6):767-91
20923728 - IEEE Trans Biomed Eng. 2011 Jan;58(1):132-43
21439324 - J Neurosci Methods. 2011 May 15;198(1):114-24
17601190 - IEEE Trans Neural Syst Rehabil Eng. 2007 Jun;15(2):207-16
18339453 - Neurobiol Aging. 2009 Dec;30(12):2065-79
3809365 - Psychophysiology. 1986 Sep;23(5):590-7
18179351 - Psychol Methods. 2007 Dec;12(4):399-413
2461285 - Electroencephalogr Clin Neurophysiol. 1988 Dec;70(6):510-23
22208120 - Clin EEG Neurosci. 2011 Oct;42(4):230-5
References_xml – reference: 19545601 - Neurosci Lett. 2009 Oct 2;462(1):94-8
– reference: 2461285 - Electroencephalogr Clin Neurophysiol. 1988 Dec;70(6):510-23
– reference: 12048038 - Clin Neurophysiol. 2002 Jun;113(6):767-91
– reference: 4695775 - Exp Brain Res. 1973 Mar 19;16(5):455-65
– reference: 20923728 - IEEE Trans Biomed Eng. 2011 Jan;58(1):132-43
– reference: 10731765 - Psychophysiology. 2000 Mar;37(2):127-52
– reference: 17601190 - IEEE Trans Neural Syst Rehabil Eng. 2007 Jun;15(2):207-16
– reference: 20509913 - Behav Brain Funct. 2010;6:28
– reference: 21439324 - J Neurosci Methods. 2011 May 15;198(1):114-24
– reference: 20214929 - Neurosci Res. 2010 Jun;67(2):172-80
– reference: 1464675 - J Clin Neurophysiol. 1992 Oct;9(4):456-79
– reference: 19041343 - J Neurosci Methods. 2009 Mar 15;177(2):488-96
– reference: 17573239 - Clin Neurophysiol. 2007 Oct;118(10):2128-48
– reference: 18179351 - Psychol Methods. 2007 Dec;12(4):399-413
– reference: 22030141 - Clin Neurophysiol. 2012 Jun;123(6):1123-30
– reference: 22143614 - J Neurol. 2012 Jun;259(6):1191-8
– reference: 18339453 - Neurobiol Aging. 2009 Dec;30(12):2065-79
– reference: 20405196 - Brain Topogr. 2010 Jun;23(2):180-5
– reference: 22208120 - Clin EEG Neurosci. 2011 Oct;42(4):230-5
– reference: 3809365 - Psychophysiology. 1986 Sep;23(5):590-7
SSID ssj0031790
Score 2.1687748
Snippet Brain-computer interfaces (BCIs) that detect event-related potentials (ERPs) rely on classification schemes that are vulnerable to latency jitter, a phenomenon...
SourceID proquest
pubmed
SourceType Aggregation Database
Index Database
StartPage 016006
SubjectTerms Adolescent
Adult
Aged
Brain - physiology
Brain-Computer Interfaces - classification
Brain-Computer Interfaces - standards
Female
Humans
Male
Middle Aged
Predictive Value of Tests
Reaction Time - physiology
Young Adult
Title Classifier-based latency estimation: a novel way to estimate and predict BCI accuracy
URI https://www.ncbi.nlm.nih.gov/pubmed/23234797
https://www.proquest.com/docview/1317855427
Volume 10
WOSCitedRecordID wos000316245500008&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText
inHoldings 1
isFullTextHit
isPrint
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV3LS8MwGA_qPHjxNR_zRQTxFpakj7ReZA6HHhw7ONhtfE1SGGg7t27S_94kbfUkCF56SAkN6Zfv--V7_RC6kSIEoOARBgDEj7QiIM25UgYtKBZLyZlyZBNiOIwmk3hUO9yWdVploxOdola5tD7yLvMsj3zgc3E__yCWNcpGV2sKjU3U8gyUsVItJt9RBM92n6oKIhkx0Jk3FTzm0lePhdSojS7r2j5rNPwdZTprM9j77zr30W6NM3GvEowDtKGzQ9TuZeaO_V7iW-wyP51LvY3GjhlzlhoLSaxVU_gNLJQusW3BUdU23mHAWb7Wb_gTSlzkzSuNIVN4vrDxngI_9J8xSLlagCyP0Hjw-Np_IjXdApF-FBYk4TpyHd2UEknMPACphPZ8miYhMDBASikVx2moRRwGqZcmAeWapYmUNIgixY_RVpZn-hRhQXlCPS649qkvUmUmUV_6TJtPgIx4B103mzc14mxjFJDpfLWc_mxfB51Uf2A6r_puTA34s3Wv4uwPs8_RDnfEFTbx5AK1UnOY9SXalutitlxcOTkxz-Ho5QvTYMcK
linkProvider ProQuest
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Classifier-based+latency+estimation%3A+a+novel+way+to+estimate+and+predict+BCI+accuracy&rft.jtitle=Journal+of+neural+engineering&rft.au=Thompson%2C+David+E&rft.au=Warschausky%2C+Seth&rft.au=Huggins%2C+Jane+E&rft.date=2013-02-01&rft.issn=1741-2552&rft.eissn=1741-2552&rft.volume=10&rft.issue=1&rft.spage=016006&rft_id=info:doi/10.1088%2F1741-2560%2F10%2F1%2F016006&rft.externalDBID=NO_FULL_TEXT
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1741-2552&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1741-2552&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1741-2552&client=summon