HOME > Detail View

Detail View

Channel selection method based on relevance score for EEG BCI with high-gamma band

Channel selection method based on relevance score for EEG BCI with high-gamma band

Material type
학위논문
Personal Author
이진우, 李鎭宇
Title Statement
Channel selection method based on relevance score for EEG BCI with high-gamma band / Jinwoo Lee
Publication, Distribution, etc
Seoul :   Graduate School, Korea University,   2021  
Physical Medium
iii, 29장 : 도표 ; 26 cm
기타형태 저록
Channel Selection Method based on Relevance Score for EEG BCI with High-gamma Band   (DCOLL211009)000000235777  
학위논문주기
학위논문(석사)-- 고려대학교 대학원: 컴퓨터·전파통신공학과, 2021. 2
학과코드
0510   6D36   1122  
General Note
지도교수: 정원주  
Bibliography, Etc. Note
참고문헌: 장 24-29
이용가능한 다른형태자료
PDF 파일로도 이용가능;   Requires PDF file reader(application/pdf)  
비통제주제어
EEG BCI , Channel selection method,,
000 00000nam c2200205 c 4500
001 000046071899
005 20210326141532
007 ta
008 201221s2021 ulkd bmAC 000c eng
040 ▼a 211009 ▼c 211009 ▼d 211009
085 0 ▼a 0510 ▼2 KDCP
090 ▼a 0510 ▼b 6D36 ▼c 1122
100 1 ▼a 이진우, ▼g 李鎭宇
245 1 0 ▼a Channel selection method based on relevance score for EEG BCI with high-gamma band / ▼d Jinwoo Lee
260 ▼a Seoul : ▼b Graduate School, Korea University, ▼c 2021
300 ▼a iii, 29장 : ▼b 도표 ; ▼c 26 cm
500 ▼a 지도교수: 정원주
502 0 ▼a 학위논문(석사)-- ▼b 고려대학교 대학원: ▼c 컴퓨터·전파통신공학과, ▼d 2021. 2
504 ▼a 참고문헌: 장 24-29
530 ▼a PDF 파일로도 이용가능; ▼c Requires PDF file reader(application/pdf)
653 ▼a EEG BCI ▼a Channel selection method
776 0 ▼t Channel Selection Method based on Relevance Score for EEG BCI with High-gamma Band ▼w (DCOLL211009)000000235777
900 1 0 ▼a Lee, Jin-woo, ▼e
900 1 0 ▼a 정원주, ▼g 鄭原周, ▼e 지도교수
900 1 0 ▼a Chung, Woon-zoo, ▼e 지도교수
945 ▼a KLPA

Electronic Information

No. Title Service
1
Channel selection method based on relevance score for EEG BCI with high-gamma band (13회 열람)
View PDF Abstract Table of Contents

Holdings Information

No. Location Call Number Accession No. Availability Due Date Make a Reservation Service
No. 1 Location Science & Engineering Library/Stacks(Thesis)/ Call Number 0510 6D36 1122 Accession No. 123066022 Availability Available Due Date Make a Reservation Service B M
No. 2 Location Science & Engineering Library/Stacks(Thesis)/ Call Number 0510 6D36 1122 Accession No. 123066023 Availability Available Due Date Make a Reservation Service B M

Contents information

Abstract

This paper presents a novel method of MI-relevant channel selection based on relevance score for electroencephalogram(EEG) brain computer interface(BCI) with high-gamma band. The proposed method performs band-pass filtering of EEG data in mu, beta and high-gamma bands, and the band-pass filtered EEG data is transformed into spectrogram by short time fourier transform (STFT). For our convolutional neural networks (CNNs), the spectrogram is used as input data.  Unlike the existing channel selection method, which selects task-relevant channels based on a correlation coefficient or a CSP-based algorithm, the proposed method applies an explainable deep learning Layer-wise Relevance Propagation(LRP) algorithm to CNN to select channels with high contribution. The simulation was conducted through the BCI Competition III dataset IVa. And as a result of the experiment, it can be seen that the performance is improved compared to the existing methods.

Table of Contents

Contents

Chapter 1. Introduction………………………………………………………………….…....1
Chapter 2. Method ....……………………………………………..……………………….....7
2.1.	System Model………………….….………....…………………………....7
2.2.	Preprocessing ……………………………………………………………..8
2.3.	Training with Convolutional Neural Networks ………………...………..10
2.4.	Layer-wise Relevance Propagation ……………………..……………….11
2.5.	Channel Selection based on Relevance Score ....…..…………………….15
2.6.	Characteristics of brain waves according to Frequency band .…………..18
Chapter 3. Data and Experiment .…..………………………….............................................20
3.1.	Data Description ....……………….………....…………………………...20
3.2.	Data Processing ………………………………………………………….20
Chapter 4. Result and Discussion ……………………..…………………………………....22
4.1.	Performance Comparison for BCI Competition III dataset IVa ….……...22
Chapter 5. Conclusion ……………………………………….………………………….…..23
References ....………………………………………………………………………………. 24