A novel scheme for the validation of an automated classification method for epileptic spikes by comparison with multiple observers.

Niraj K. Sharma; Carlos Pedreira; Maria Centeno; Umair J. Chaudhary; Tim Wehner; Lucas G.S. França; Tinonkorn Yadee; Teresa Murta; Marco Leite; Sjoerd B. Vos; Sebastien Ourselin; Beate Diehl; Louis Lemieux

doi:10.1016/j.clinph.2017.04.016

A novel scheme for the validation of an automated classification method for epileptic spikes by comparison with multiple observers.

Niraj K. Sharma^*, Carlos Pedreira, Maria Centeno, Umair J. Chaudhary, Tim Wehner, Lucas G.S. França, Tinonkorn Yadee, Teresa Murta, Marco Leite, Sjoerd B. Vos, Sebastien Ourselin, Beate Diehl, Louis Lemieux

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

9 Citations (Scopus)

5 Downloads (Pure)

Abstract

Objective
To validate the application of an automated neuronal spike classification algorithm, Wave_clus (WC), on interictal epileptiform discharges (IED) obtained from human intracranial EEG (icEEG) data.

Method
Five 10-min segments of icEEG recorded in 5 patients were used. WC and three expert EEG reviewers independently classified one hundred IED events into IED classes or non-IEDs. First, we determined whether WC-human agreement variability falls within inter-reviewer agreement variability by calculating the variation of information for each classifier pair and quantifying the overlap between all WC-reviewer and all reviewer-reviewer pairs. Second, we compared WC and EEG reviewers’ spike identification and individual spike class labels visually and quantitatively.

Results
The overlap between all WC-human pairs and all human pairs was >80% for 3/5 patients and >58% for the other 2 patients demonstrating WC falling within inter-human variation. The average sensitivity of spike marking for WC was 91% and >87% for all three EEG reviewers. Finally, there was a strong visual and quantitative similarity between WC and EEG reviewers.

Conclusions
WC performance is indistinguishable to that of EEG reviewers’ suggesting it could be a valid clinical tool for the assessment of IEDs.

Significance
WC can be used to provide quantitative analysis of epileptic spikes.

Original language	English
Pages (from-to)	1246-1254
Number of pages	9
Journal	Clinical Neurophysiology
Early online date	4 May 2017
DOIs	https://doi.org/10.1016/j.clinph.2017.04.016
Publication status	Published - 1 Jul 2017
Externally published	Yes

Keywords

Interictal spike classification
Intracranial EEG
Automated spike classification
Information theory

Access to Document

10.1016/j.clinph.2017.04.016Licence: CC BY

1-s2.0-S1388245717301621-mainFinal published version, 1.21 MBLicence: CC BY

Cite this

Sharma, N. K., Pedreira, C., Centeno, M., Chaudhary, U. J., Wehner, T., França, L. G. S., Yadee, T., Murta, T., Leite, M., Vos, S. B., Ourselin, S., Diehl, B., & Lemieux, L. (2017). A novel scheme for the validation of an automated classification method for epileptic spikes by comparison with multiple observers. Clinical Neurophysiology, 1246-1254. https://doi.org/10.1016/j.clinph.2017.04.016

@article{7bd27711455d41fb93ff81b0228b84c3,

title = "A novel scheme for the validation of an automated classification method for epileptic spikes by comparison with multiple observers.",

abstract = "Objective To validate the application of an automated neuronal spike classification algorithm, Wave\_clus (WC), on interictal epileptiform discharges (IED) obtained from human intracranial EEG (icEEG) data. Method Five 10-min segments of icEEG recorded in 5 patients were used. WC and three expert EEG reviewers independently classified one hundred IED events into IED classes or non-IEDs. First, we determined whether WC-human agreement variability falls within inter-reviewer agreement variability by calculating the variation of information for each classifier pair and quantifying the overlap between all WC-reviewer and all reviewer-reviewer pairs. Second, we compared WC and EEG reviewers{\textquoteright} spike identification and individual spike class labels visually and quantitatively. Results The overlap between all WC-human pairs and all human pairs was >80\% for 3/5 patients and >58\% for the other 2 patients demonstrating WC falling within inter-human variation. The average sensitivity of spike marking for WC was 91\% and >87\% for all three EEG reviewers. Finally, there was a strong visual and quantitative similarity between WC and EEG reviewers. Conclusions WC performance is indistinguishable to that of EEG reviewers{\textquoteright} suggesting it could be a valid clinical tool for the assessment of IEDs. Significance WC can be used to provide quantitative analysis of epileptic spikes.",

keywords = "Interictal spike classification, Intracranial EEG, Automated spike classification, Information theory",

author = "Sharma, \{Niraj K.\} and Carlos Pedreira and Maria Centeno and Chaudhary, \{Umair J.\} and Tim Wehner and Fran{\c c}a, \{Lucas G.S.\} and Tinonkorn Yadee and Teresa Murta and Marco Leite and Vos, \{Sjoerd B.\} and Sebastien Ourselin and Beate Diehl and Louis Lemieux",

year = "2017",

month = jul,

day = "1",

doi = "10.1016/j.clinph.2017.04.016",

language = "English",

pages = "1246--1254",

journal = "Clinical Neurophysiology",

issn = "1388-2457",

publisher = "Elsevier",

}

Sharma, NK, Pedreira, C, Centeno, M, Chaudhary, UJ, Wehner, T, França, LGS, Yadee, T, Murta, T, Leite, M, Vos, SB, Ourselin, S, Diehl, B & Lemieux, L 2017, 'A novel scheme for the validation of an automated classification method for epileptic spikes by comparison with multiple observers.', Clinical Neurophysiology, pp. 1246-1254. https://doi.org/10.1016/j.clinph.2017.04.016

TY - JOUR

T1 - A novel scheme for the validation of an automated classification method for epileptic spikes by comparison with multiple observers.

AU - Sharma, Niraj K.

AU - Pedreira, Carlos

AU - Centeno, Maria

AU - Chaudhary, Umair J.

AU - Wehner, Tim

AU - França, Lucas G.S.

AU - Yadee, Tinonkorn

AU - Murta, Teresa

AU - Leite, Marco

AU - Vos, Sjoerd B.

AU - Ourselin, Sebastien

AU - Diehl, Beate

AU - Lemieux, Louis

PY - 2017/7/1

Y1 - 2017/7/1

N2 - Objective To validate the application of an automated neuronal spike classification algorithm, Wave_clus (WC), on interictal epileptiform discharges (IED) obtained from human intracranial EEG (icEEG) data. Method Five 10-min segments of icEEG recorded in 5 patients were used. WC and three expert EEG reviewers independently classified one hundred IED events into IED classes or non-IEDs. First, we determined whether WC-human agreement variability falls within inter-reviewer agreement variability by calculating the variation of information for each classifier pair and quantifying the overlap between all WC-reviewer and all reviewer-reviewer pairs. Second, we compared WC and EEG reviewers’ spike identification and individual spike class labels visually and quantitatively. Results The overlap between all WC-human pairs and all human pairs was >80% for 3/5 patients and >58% for the other 2 patients demonstrating WC falling within inter-human variation. The average sensitivity of spike marking for WC was 91% and >87% for all three EEG reviewers. Finally, there was a strong visual and quantitative similarity between WC and EEG reviewers. Conclusions WC performance is indistinguishable to that of EEG reviewers’ suggesting it could be a valid clinical tool for the assessment of IEDs. Significance WC can be used to provide quantitative analysis of epileptic spikes.

AB - Objective To validate the application of an automated neuronal spike classification algorithm, Wave_clus (WC), on interictal epileptiform discharges (IED) obtained from human intracranial EEG (icEEG) data. Method Five 10-min segments of icEEG recorded in 5 patients were used. WC and three expert EEG reviewers independently classified one hundred IED events into IED classes or non-IEDs. First, we determined whether WC-human agreement variability falls within inter-reviewer agreement variability by calculating the variation of information for each classifier pair and quantifying the overlap between all WC-reviewer and all reviewer-reviewer pairs. Second, we compared WC and EEG reviewers’ spike identification and individual spike class labels visually and quantitatively. Results The overlap between all WC-human pairs and all human pairs was >80% for 3/5 patients and >58% for the other 2 patients demonstrating WC falling within inter-human variation. The average sensitivity of spike marking for WC was 91% and >87% for all three EEG reviewers. Finally, there was a strong visual and quantitative similarity between WC and EEG reviewers. Conclusions WC performance is indistinguishable to that of EEG reviewers’ suggesting it could be a valid clinical tool for the assessment of IEDs. Significance WC can be used to provide quantitative analysis of epileptic spikes.

KW - Interictal spike classification

KW - Intracranial EEG

KW - Automated spike classification

KW - Information theory

UR - https://www.scopus.com/pages/publications/85019634498

U2 - 10.1016/j.clinph.2017.04.016

DO - 10.1016/j.clinph.2017.04.016

M3 - Article

C2 - 28531810

SN - 1388-2457

SP - 1246

EP - 1254

JO - Clinical Neurophysiology

JF - Clinical Neurophysiology

ER -

A novel scheme for the validation of an automated classification method for epileptic spikes by comparison with multiple observers.

Abstract

Keywords

Access to Document

Other files and links

Cite this