Content uploaded by Christoph Guger
Author content
All content in this area was uploaded by Christoph Guger on Apr 21, 2020
Content may be subject to copyright.
CURRENT AND
FUTURE BRAIN-
COMPUTER
INTERFACES
Christoph Guger
g.tec medical engineering
3017 persons registered
Biggest BCI and Neurotechnology Meeting ever !!
Certificate available on Friday
Companies
•Guger Technologies OG, Graz, Austria
•g.tec medical engineering GmbH, Schiedlberg, Austria
•g.tec medical engineering Spain SL, Barcelona, Spain
•g.tec neurotechnology USA Inc., Albany, USA
•g.tec neurotechnology Hong Kong Inc., HK
Research Projects
•H2020 Eurostars: seizureAI –epilepsy monitoring and deep learning
with EEG
•H2020 Eurostars: ComaAid –combination of EEG and fNIRS for DOC
patients
•H2020 Eurostars: EEG-DDS –combination of EEG with decision support
system
•FFG: BrainGait –combination of BCIs for rehabilitation with treadmills
•H2020: MultiSense –combination of DBS with physiology
•FFG: recoveriX-Leg –lower limb stroke rehabilitation with BCI
•H2020: Rhumbo –magnetic stimulation of the cortex
Research Projects
•H2020: HOPE –high frequency oscillation detection in epilepsy
•H2020 ITN: Pro-Gait –combination of BCIs with exoskeletons
•H2020 ITN: DOCMA –international training network for BCIs and
DOC
•Marie-Curie: MoveAgain –combination of recoveriX with TMS, tDCS
•H2020: Astonish –fNIRS and EEG
•H2020 Eurostars: ComAlert –coma prediction
•EC Flagship: Graphene –development of Graphene electrodes
•BeMagic –magnetic stimulation of the brain
BRAIN-COMPUTER INTERFACE
BRAIN-COMPUTER INTERFACE
BRAIN-COMPUTER INTERFACE
“A system for controlling a device e.g. computer,
wheelchair or a neuroprostheses by human intention which
does not depend on the brain’s normal output pathways of
peripheral nerves and muscles” [Wolpaw et al., 2002].
Subject/
Patient
Brain-
Computer
Interface
Device
Feedback
EEG/
ECoG control signal
32/16/8 channel wireless EEG
3-axis accelerometer
24 Bit accuracy at 250/500 Hz
A new benchmark in usability
Active electrode technology
Waterproof device
Contactless charging
2.4 GHz digital transmission
BCI PRINCIPLES
- Slow cortical potentials (anticipation tasks)
- Steady-State Evoked potentials (SSVEP, SSSEP)
- Event-related, non-phase-locked changes of oscillatory activity
ERD/ERS (motor imagery tasks)
- Evoked potentials (focus on attention task)
- Code based Evoked potentials (focus on a code)
BRAIN-COMPUTER INTERFACE
BRAIN-COMPUTER INTERFACE
BRAIN-COMPUTER INTERFACE
BRAIN-COMPUTER INTERFACE
Physiological background
Imagination of hand movement causes an ERD which is used to classify the
side of movement. The desynchronization occurs in motor and related areas of
the brain.
Left hand
movement Right hand
movement
C4
GND
REF
RIGHT
C3
•Rehabilitation Hospital of Iasi, Romania
•Training video: recoveriX1
9 Hole Peg Test
Date:
2014
Left
hand
Falls Right
hand
Falls
24-06 31” 01’5’’ 0
26-06 32” 054” 0
29-06 32” 045” 0
2-07 31” 042” 0
6-09 31” 042” 0
9-09 29” 038” 0
12-09 29” 034” 0
15-09 29” 030” 0
11-01 29” 030” 0
recoveriX Results
ID: S.E. Age: 62 Female Affected Side: Right Time since stroke: 12 months
recoveriX Results
ID: S.E. Age: 62 Female Affected Side: Right Time since stroke: 12 months
17,1
13,6 11,6
15,7
21,3
15,5
10,2
21,3
13,8 12,5 10 9,8 12,3 14,1 12,3
21,6
15,5
19,3 19,8
12,9
20,2
12,5
5 5
11,3 12,5
8,8
5
15
7,5 8,8 6,3 53,8
7,5 8,8
17,5
10 8,8
13,8
8,8 11,3
0
5
10
15
20
25
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21
Mean Error Min. Error
Session 1 Session 13
After training:
clear left/right
difference
faster minimum
recoveriX Results
ID: S.E. Age: 62 Female Affected Side: Right Time since stroke: 12 months
Motor imagery works very well with right, left
hand and foot, but it is difficult to get more
degrees of freedom
Basic spectral changes with movement
credits: Kai Miller
Real-time representation of cortical
activity
Rapid Functional ECoG Mapping
P300 –Approach (EEG)
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O PQ R
S T U V W X
YZ0 1 2 3
45 6 7 8 9
visual stimulation Time (ms)
Amplitude (µV)
78
Optic nerve
P300
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
P300
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
AB C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
P300
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P QR
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
P300
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
MN O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
P300
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V WX
Y Z 0 1 2 3
4 5 6 7 8 9
www.gtec.at
The 6 x 6 matrix speller
Target:15
µV
Non-target: 1 µV
Letter W Presentation
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7
-10
-8
-6
-4
-2
0
2
4
6
8
10
time [s]
[µV]
P300
P300
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V WX
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
P300
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V WX
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
ABC D E F
GHI J K L
MNO P Q R
STU V W X
YZ0 1 2 3
456 7 8 9
P300
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V WX
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
AB C D E F
GH I J K L
MN O P Q R
ST U V W X
YZ 0 1 2 3
45 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
P300
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V WX
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
AB C D E F
GH I J K L
MN O P Q R
ST U V W X
YZ 0 1 2 3
45 6 7 8 9
P300
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V WX
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
A B C D E F
G H I J K L
M N O P Q R
S T U V W X
Y Z 0 1 2 3
4 5 6 7 8 9
P300
P300
Everybody achieved 100 % accuracy
17 subjects
4 tasks
Standard flashing
„Einstein“ (black/white)
„Einstein“ (color)
Face speller
FACE SPELLER
Disorders of Consciousness
Fact: 43% of patients diagnosed as vegetative are reclassified as (at least) minimally conscious
when investigated by expert teams
Disorders of Consciousness
Fact: 43% of patients diagnosed as vegetative are reclassified as (at least) minimally conscious
when investigated by expert teams
AUDITORY RESULTS
YES/NO ANSWERS
YES/NO ANSWERS
www.gtec.at
SSVEP
7 Hz
Steady-State-Evoked Potentials
Higher
Frequency
(e.g. 17 Hz)
Lower
Frequency
(e.g. 14 Hz)
EEG Power Spectrum EEG Power Spectrum
World of Warcraft
VR/AR
Hand Movement Task
4 spatial filters per class
→12 features for movement
discrimination
maximizing patterns minimizing patterns
1
-1
0
Hand Movement Task
Movement: Online classification accuracy
Imagination: Online classification accuracy
BCI-ECoG setup Classification accuracy
FOLLOW US ON INSTRAGRAM
@unicorn_brain_interface
www.unicorn-bi.com
FOLLOW US ON FACEBOOK
@gtecmedicalengineering
www.gtec.at
FOLLOW US ON TWITTER
@gtec_BCI
FOLLOW US ON YOUTUBE
youtube.com/c/gtecmedicalengineering
www.gtec.at
BIOSIGNAL
AMPLIFIERS