en-es
en
0.25
0.5
0.75
1.25
1.5
1.75
2
Oscillatory EEG-based BCI design: signal processing and more
Published on Apr 03, 20144222 Views
This lecture proposes an accessible introduction to the design of Brain-Computer Interfaces (BCI) based on oscillatory EEG activity (e.g., motor imagery), notably from a signal processing point of vie
Related categories
Chapter list
Oscillatory EEG-based BCI design: signal processing and more00:00
Oscillatory activity-based BCI00:53
Example: Motor Imagery (MI)01:37
Architecture of a BCI04:25
This talk main focus04:53
Lecture Outline05:01
Basic oscillatory activity-based BCI design06:04
Oscillatory EEG-based BCI design:apattern recognition approach06:09
Example: Features for Motor Imagery-based BCI06:51
Band power features - 109:36
Band power features - 210:56
Basic design for left and right hand motor imagery-based BCI11:32
That’s all? It’s that simple?12:23
Basic design performance examples13:24
Using more channels16:06
Spatial Filters & Common Spatial Patterns17:00
(Linear) Spatial Filtering17:03
Some basic spatial filters18:17
Basic spatial filters performance examples19:30
More advanced spatial filters20:59
Inverse Solutions21:33
Supervised spatial filtering:Common Spatial Patterns (CSP) informally…22:33
CSP formally23:42
CSP in action - 127:27
CSP in action - 229:01
CSP performance examples30:11
Pros and cons of CSP31:11
Common Spatial Patterns Extensions32:31
Towards a more robust BCI?32:46
Regularized CSP (RCSP)33:54
What prior knowledge to use? Spatial knowledge to deal with noise36:03
Spatial filters obtained38:41
Regularization terms to deal with non-stationarities40:00
Combining multiple regularization terms41:42
Regularization terms to reduce calibration time42:14
Evaluation43:39
Sparse CSP for channel selection45:01
Using a-priori knowledge for CSP47:26
Spatial filters for relating EEG band power to a target variable48:11
SPoC: Source Power Comodulation49:45
SPoC example51:11
In short, spatial filters are really useful…But do we really need them?53:58
Classifying Covariance matrices directly54:07
Summary of CSP spatial filters extensions55:24
OptimizingSpectral Filtering55:56
Using Subject-specific frequency bands56:34
Optimizing Spatio-Spectral Filters for BCI58:32
The Filter Bank CSP (FBCSP)59:28
FBCSP Results01:01:48
Optimizing temporal filters for BCI - 101:02:26
Optimizing temporal filters for BCI - 201:04:18
Discriminative Filter Bank CSP (DFBCSP)01:05:57
DFBCSP results01:06:54
Alternative Features for Oscillatory activity based BCI01:07:07
Feedback & User training01:07:12
A BCI is a co-adaptive system01:07:58
BCI skills01:09:02
What does classical BCI training looks like?01:10:19
Some remarks on standard BCI feedback01:10:55
On BCI standard user training approaches01:11:57
Improving BCI feedback01:12:57
Improving BCI training environment01:15:07
Improving training tasks01:17:51
The point of view of instructional design01:20:33
Summary on user training for BCI01:23:02
Summary and Conclusion01:23:50
Summary on oscillatory activity-based BCI-design01:23:58
Some related open research challenges01:24:58
Take home messages for oscillatory activity EEG-based BCI design01:25:55
Thank you for your attention!01:26:25