Sign inSign up
Spatio-Spectral Filter Optimization in a Bayesian Framework for Single-Trial EEG Classification in Brain-Computer Interface thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Spatio-Spectral Filter Optimization in a Bayesian Framework for Single-Trial EEG Classification in Brain-Computer Interface

Published on Dec 03, 20123848 Views

Seong-Whan Lee

There are two challenging problems in classifying a single-trial EEG of motor imagery. One is spectral filter optimization - The frequency bands, in which ERD/ERS patterns reflect activation and deact

BBCI Workshop 2012 - Berlin

Related categories

NeuroscienceNeurotechnology

Chapter list

Spatio-Spectral Filter Optimization in a Bayesian Framework for EEG-based Brain-Computer Interfaces00:00
Brain-Computer Interface (BCI)00:30
Prevalent Steps in Motor Imagery Classification03:29
Related Work04:30
A Probabilistic Bayesian Approach05:14
A Novel Bayesian Framework06:31
Main Contributions07:45
Mathematical Formulation (1)08:29
Mathematical Formulation (2)09:38
Posterior Estimation10:12
Posterior Estimation (cont.)11:05
Likelihood Computation11:40
Mutual Information12:30
A Novel Bayesian Framework (recap.)13:02
Spectrally Weighted Classification13:21
Three Public Datasets14:14
Preprocessing and Hyperparameter Setting14:48
How Many Particles in pdf Estimation15:37
Performance on TU Berlin Dataset16:27
Performance on TU Berlin Dataset (cont.)16:45
Performance on BCI Competition III-IVa17:15
Performance on BCI Competition IV-IIa17:55
Performance on BCI Competition IV-Iia (cont.)17:57
Statistical Significance Test17:59
Statistical Significance Test (cont.)18:26
Conclusions18:39
Further Issues19:59
Thank you for your attention20:49