Back to search

University of South Florida

EEG Classifier Validation Methods for Neuroprosthetic Hand Development

Abstract

dc:description

To date, many challenges have been reported in the development of neuroprosthetic hands using EEG and neural signals. In this study, we report the results of a literature review on Brain Computer Interface (BCI) technology, an investigation of estimation methods using applications in MATLAB, and the results of Electroencephalography (EEG) classification to assist in the development of neural prosthetic hands using biological signals such as EEG. Confusion Matrix was created using Motor Imagery (MI) data as the predictive value, and the average accuracy of more than 90% was obtained for the K-Nearest Neighbor (KNN) and decision tree method. The results were evaluated with 5-Cross Validation and discussed with Confusion Matrix. The results suggest that misclassification may occur in multiclass classification with more than two types of directions including bimanual movements, as in this study, because the MI of dominant and non-dominant hands are significantly different.

Degree

thesis:*
Name thesis:degree_name
MS in Biomedical Engineering (M.S.B.E.)
Level thesis:degree_level
Thesis
Discipline thesis:degree_discipline
Engineering
Grantor
University of South Florida
Year dc:date
2023

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Yamauchi, Keigo
Contributors dc:contributor
  • Stephen E. Saddow, Ph.D.
  • Mark Jaroszeski, Ph.D.
  • Robert Frisina, Jr., Ph.D.

Subjects

dc:subject × 6

Identifiers

dc:identifier.*
Repository record dc:identifier
https://digitalcommons.usf.edu/etd/9947

Chain of custody

source
Harvested from
University of South Florida
Base URL
digitalcommons.usf.edu/do/oai/
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
citation

Yamauchi, Keigo. EEG Classifier Validation Methods for Neuroprosthetic Hand Development. Thesis thesis, University of South Florida, 2023. https://digitalcommons.usf.edu/etd/9947