Parameterization and classification of VEP after selective cone excitation

P. Bessler*, J. Rieger, S. Klee, J. Haueisen

*Corresponding author for this work

Research output: Contribution to book/report/conference proceedingContribution to conference proceeding

Abstract

The purpose of the work was the objective analysis of the cortical response signals after selective colour channel stimulation. Using the parameters of the visual evoked potential (VEP) we aimed to classify the separate colour channels as well as glaucoma patients and volunteers. On the basis of via the silent substitution technique (SST) flash stimulation, a selective S-cone stimulation and selective L- and M-cone stimulation was performed. 102 healthy subjects and 39 glaucoma patients were studied. The VEPs were recorded, pre-processed and analyzed. Based on the response signals, parameters were extracted. Using the parameters, a classification was done by the use of the binary logistic regression. The S-cone response and the LM-cone response could be differentiated with accuracy about 80%. Volunteers and glaucoma patients of the test group could be classified with overall accuracy of 85% at S-cone response and 76% at LM-cone response. The peak latency and reconstruction error seems to be a meaningful parameter for an effective classification. Good classification results could be achieved for the separate colour channels as well as for volunteers and glaucoma patients. The classification of glaucoma patients was clearly better by the use of the S-cone responses than by the use of LM-cone responses. Larger set of signals, especially from glaucoma patients, would probably improve the classification results.

Original languageEnglish
Title of host publicationWorld Congress on Medical Physics and Biomedical Engineering
Subtitle of host publicationBiomedical Engineering for Audiology, Ophthalmology, Emergency and Dental Medicine
PublisherSpringer Verlag
Pages168-171
Number of pages4
Edition11
ISBN (Print)9783642038907
DOIs
Publication statusPublished - 2009
Externally publishedYes
EventWorld Congress on Medical Physics and Biomedical Engineering: Biomedical Engineering for Audiology, Ophthalmology, Emergency and Dental Medicine - Munich, Germany
Duration: 07 Sept 200912 Sept 2009

Publication series

NameIFMBE Proceedings
Number11
Volume25
ISSN (Print)1680-0737

Conference

ConferenceWorld Congress on Medical Physics and Biomedical Engineering: Biomedical Engineering for Audiology, Ophthalmology, Emergency and Dental Medicine
Country/TerritoryGermany
CityMunich
Period07.09.200912.09.2009

Keywords

  • Binary logistic regression
  • Classification
  • Cones
  • SST
  • VEP

ASJC Scopus subject areas

  • Bioengineering
  • Biomedical Engineering

Fingerprint

Dive into the research topics of 'Parameterization and classification of VEP after selective cone excitation'. Together they form a unique fingerprint.

Cite this