A quality metric to improve wrapper feature selection in multiclass subject invariant brain computer interfaces

MOspace/Manakin Repository

Breadcrumbs Navigation

A quality metric to improve wrapper feature selection in multiclass subject invariant brain computer interfaces

Please use this identifier to cite or link to this item: http://hdl.handle.net/10355/14510

[-] show simple item record

dc.contributor.advisor Derakhshani, Reza en
dc.contributor.author Sherwood, Jesse
dc.date.accessioned 2012-06-05T14:39:40Z
dc.date.available 2012-06-05T14:39:40Z
dc.date.issued 2012-06-05
dc.date.submitted 2011 Summer en
dc.identifier.uri http://hdl.handle.net/10355/14510
dc.description Title from PDF of title page, viewed on June 5, 2012 en
dc.description Dissertation advisor: Reza Derakhshani en
dc.description Vita en
dc.description Includes bibliographical references (p. 116-129) en
dc.description Thesis (Ph.D.)--School of Computing and Engineering. University of Missouri--Kansas City, 2012 en
dc.description.abstract Brain computer interface systems based on electroencephalograph (EEG) signals have limitations which challenge their application as a practical device for general use. The signal features generated by the brain states we wish to detect possess a high degree of inter-subject and intra-subject variation. Additionally, these features usually exhibit a low variation across each of the target states. Collection of EEG signals using low resolution, non-invasive scalp electrodes further degrades the spatial resolution of these signals. The majority of brain computer interface systems to date require extensive training prior to use by each individual user. The discovery of subject invariant features could reduce or even eliminate individual training requirements. To obtain suitable subject invariant features requires search through a high dimension feature space consisting of combinations of spatial, spectral and temporal features. Poorly separable features can prevent the search from converging to a usable solution as a result of degenerate classifiers. In such instances the system must detect and compensate for degenerate classifier behavior. This dissertation presents a method to accomplish this search using a wrapper architecture comprised of a sequential forward floating search algorithm coupled with a support vector machine classifier. This is successfully achieved by the introduction of a scalar Quality (Q)-factor metric, calculated from the ratio of sensitivity to specificity of the confusion matrix. This method is successfully applied to a multiclass subject independent BCI using 10 untrained subjects performing 4 motor tasks. en_US
dc.description.tableofcontents Introduction to brain computer interface systems -- Historical perspective and state of the art -- Experimental design -- Degeneracy in support vector machines -- Discussion of research -- Results -- Conclusion -- Appendix A. Information transfer rate -- Appendix B. Additional surface plots for individual tasks and subjects en
dc.format.extent xi, 131 pages en
dc.language.iso en_US en_US
dc.publisher University of Missouri--Kansas City en
dc.subject.lcsh Brain-computer interfaces en
dc.subject.other Dissertation -- University of Missouri--Kansas City -- Engineering en
dc.subject.other Dissertation -- University of Missouri--Kansas City -- Computer science en
dc.title A quality metric to improve wrapper feature selection in multiclass subject invariant brain computer interfaces en_US
dc.type Thesis en_US
thesis.degree.discipline Electrical and Computer Engineering and Telecommunications and Computer Networking en
thesis.degree.grantor University of Missouri--Kansas City en
thesis.degree.name Ph.D. en
thesis.degree.level Doctoral en


This item appears in the following Collection(s)

[-] show simple item record