Detection of epileptiform spikes in the EEG using a patient-independent neural network

Kerry Wilson; W. Robert S. Webber; Ronald P. Lesser; Robert S. Fisher; Russell C. Eberhart; Roy W. Dobbins

Detection of epileptiform spikes in the EEG using a patient-independent neural network

Kerry Wilson, W. Robert S. Webber, Ronald P. Lesser, Robert S. Fisher, Russell C. Eberhart, Roy W. Dobbins

School of Medicine

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

12 Scopus citations

Abstract

An offline neural network that successfully detects spikes when trained on multiple patients selected from a database of electroencephalogram (EEG) records with spikes marked by experienced electroencephalographers has been developed. This spike detector uses a simple threshold detector to identify potential spikes that appear on four-channel bipolar chains within the montage, and then passes waveform parameters to a three-layer neural network for second-level detection. Results obtained for the neural network with output thresholds arbitrarily set of 0.5 have yielded sensitivities averaging 74% and selectivities averaging 54%. While the selectivities for these trials were only fair, it is noted that substantial improvements could be achieved by raising the output thresholds.

Original language	English (US)
Title of host publication	Proc 4 Annu Symp Comput Based Med Syst
Publisher	Publ by IEEE
Pages	264-271
Number of pages	8
ISBN (Print)	0818621648
State	Published - Jan 1 1991
Event	Proceedings of the 4th Annual Symposium on Computer-Based Medical Systems - Duration: May 12 1991 → May 14 1991

Publication series

Name	Proc 4 Annu Symp Comput Based Med Syst

Other

Other	Proceedings of the 4th Annual Symposium on Computer-Based Medical Systems
Period	5/12/91 → 5/14/91

ASJC Scopus subject areas

General Engineering

Cite this

@inproceedings{53a3edfc33d149a49fa43817da09df29,

title = "Detection of epileptiform spikes in the EEG using a patient-independent neural network",

abstract = "An offline neural network that successfully detects spikes when trained on multiple patients selected from a database of electroencephalogram (EEG) records with spikes marked by experienced electroencephalographers has been developed. This spike detector uses a simple threshold detector to identify potential spikes that appear on four-channel bipolar chains within the montage, and then passes waveform parameters to a three-layer neural network for second-level detection. Results obtained for the neural network with output thresholds arbitrarily set of 0.5 have yielded sensitivities averaging 74% and selectivities averaging 54%. While the selectivities for these trials were only fair, it is noted that substantial improvements could be achieved by raising the output thresholds.",

author = "Kerry Wilson and Webber, {W. Robert S.} and Lesser, {Ronald P.} and Fisher, {Robert S.} and Eberhart, {Russell C.} and Dobbins, {Roy W.}",

year = "1991",

month = jan,

day = "1",

language = "English (US)",

isbn = "0818621648",

series = "Proc 4 Annu Symp Comput Based Med Syst",

publisher = "Publ by IEEE",

pages = "264--271",

booktitle = "Proc 4 Annu Symp Comput Based Med Syst",

note = "Proceedings of the 4th Annual Symposium on Computer-Based Medical Systems ; Conference date: 12-05-1991 Through 14-05-1991",

}

TY - GEN

T1 - Detection of epileptiform spikes in the EEG using a patient-independent neural network

AU - Wilson, Kerry

AU - Webber, W. Robert S.

AU - Lesser, Ronald P.

AU - Fisher, Robert S.

AU - Eberhart, Russell C.

AU - Dobbins, Roy W.

PY - 1991/1/1

Y1 - 1991/1/1

N2 - An offline neural network that successfully detects spikes when trained on multiple patients selected from a database of electroencephalogram (EEG) records with spikes marked by experienced electroencephalographers has been developed. This spike detector uses a simple threshold detector to identify potential spikes that appear on four-channel bipolar chains within the montage, and then passes waveform parameters to a three-layer neural network for second-level detection. Results obtained for the neural network with output thresholds arbitrarily set of 0.5 have yielded sensitivities averaging 74% and selectivities averaging 54%. While the selectivities for these trials were only fair, it is noted that substantial improvements could be achieved by raising the output thresholds.

AB - An offline neural network that successfully detects spikes when trained on multiple patients selected from a database of electroencephalogram (EEG) records with spikes marked by experienced electroencephalographers has been developed. This spike detector uses a simple threshold detector to identify potential spikes that appear on four-channel bipolar chains within the montage, and then passes waveform parameters to a three-layer neural network for second-level detection. Results obtained for the neural network with output thresholds arbitrarily set of 0.5 have yielded sensitivities averaging 74% and selectivities averaging 54%. While the selectivities for these trials were only fair, it is noted that substantial improvements could be achieved by raising the output thresholds.

UR - http://www.scopus.com/inward/record.url?scp=0025844553&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0025844553&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0025844553

SN - 0818621648

T3 - Proc 4 Annu Symp Comput Based Med Syst

SP - 264

EP - 271

BT - Proc 4 Annu Symp Comput Based Med Syst

PB - Publ by IEEE

T2 - Proceedings of the 4th Annual Symposium on Computer-Based Medical Systems

Y2 - 12 May 1991 through 14 May 1991

ER -

Detection of epileptiform spikes in the EEG using a patient-independent neural network

Abstract

Publication series

Other

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this