Scribd
Upload
9 views

BE520Lecture04 2016

Uploaded by

George Derpl
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
9 views

BE520Lecture04 2016

Uploaded by

George Derpl
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 8

BE520 – Brain Machine Interfaces

Lecture 04 – Characteristics and Measurement of EEG


OBJECTIVES
 Correlate the impact of tangential and radial dipoles on surface EEG spatial distribution.
 Quantitatively describe the role of phase synchrony and coherence of neural activities in the
generation of field potentials.
 Describe the generation of event-related potential (ERP) and event-related oscillations (ERO).
 Identify the key features of EEG signal and the setup procedure.
 Compare the different features of EEG electrode types (active vs. passive).
 Identify the difference EEG montages.
 Compare the effect of different EEG reference selection methods and their impacts on the signal
quality.

DIPOLE DIRECTION AND SURFACE EEG DISTRIBUTION


Impact of tangential and radial dipoles on the surface EEG spatial distribution:

Radial

Tangential

A.W.L.Chiu L04 Page 1 of 8


BE520 – Brain Machine Interfaces

EEG requires synchronous dipoles


 Large EEG amplitude coincides with
times when action potential sources
are approximately _____________ -
______________ with small or zero
phase shifts, due to superposition.

 Frequency bands in EEG are associated with different normal or pathological conditions.

Frequency Band Normal Conditions Pathological Conditions


 Adult slow-wave sleep  Subcortical lesions
  Deep midline lesions

 REM sleep  Subcortical lesions


 Integrative experience  Metabolic disorders
 Hydrocephalus

 Relax / reflective  Coma


 Closing of the eyes

 Active thinking  Insomnia


 ADHD

A.W.L.Chiu L04 Page 2 of 8


BE520 – Brain Machine Interfaces

QUANTITATIVE MEASURE OF PHASE


Given a measured signal with time-varying
frequency, how can we determine the phase
information?

 Phase information obtained from spectral


analysis – ________________Transform.

 Not limited to a single frequency; could be


time varying.

Hilbert transform is great for ____________________________ signals.

 Observed signal (a real waveform) x (t)


 Hilbert transform of x (t) gives a companion signal:

^x (t )=¿
 An analytic signal (complex signal) is defined to be:
jθ ( t )
x a ( t )= x ( t )+ j x^ (t)=R ( t ) e

 Instantaneous frequency is defined as:

A.W.L.Chiu L04 Page 3 of 8


BE520 – Brain Machine Interfaces

Typically phase is expressed from - to +, but to calculate the rate of change, we need to unwrap it.

The analytic signal is calculated from a ____________________ EEG by applying the Hilbert transform.

EVENT-RELATED OSCILLATION VS. EVENT-RELATED POTENTIAL

A.W.L.Chiu L04 Page 4 of 8


BE520 – Brain Machine Interfaces

Categorizing events
Underlying these EEG signals are triggers called events (cues). Events can be created externally or
internally.

 Flashing of light by LEDs

 Switching attention

 Imagined motor movements

 Beeping sounds

 Wearable tactile stimulators

Event-related oscillations (ERO) are characterized by the ______________________________(or power)


of particular _________________________________________________.

Event-related potential (ERP) are electrical brain responses _______________________________ to


physical stimuli or behavioral responses and characterized by their ___________________________ and
___________________________ in relation to stimulus onset.

PROCEDURE FOR EEG RECORDING

Why is this called 10-20 electrode configuration?

A.W.L.Chiu L04 Page 5 of 8


BE520 – Brain Machine Interfaces

A.W.L.Chiu L04 Page 6 of 8


BE520 – Brain Machine Interfaces

Naming conventions
 Prefix letters denote general cortical region
Location Functions
Fp  Top-down _____________________;
 Internal executive control
F  ___________________ of future events;
 Determine socially acceptable behavior, and danger
C Center (not a lobe)  ---

P  Integration of _________________ information (spatial


referencing) and __________________ control
O  ___________________ information processing

T  ___________________ information processing;


 Language recognition

 Odd numbers are from the left; Even numbers are from the right; midline electrodes are labeled Z

Other configurations
System # electrodes Inter-electrode distance Coverage
10-20 21 53 – 74 mm ~50%
10-10 Up to 74 28 – 38 mm ~64%
10-5 Up to 345 22 – 31 mm ~64%

THREE TYPES OF REFERENCE SYSTEMS


Referential
A single location with "no signals" as reference.
 It is not possible to find a point on the head that is considered “infinite” distance away to be a
perfect reference.
 Reference electrode test – If the ____________________ and ____________________ are
unchanged with changes in reference location!

Most likely configurations


 _______________________________ or _________________________________.
 Cz electrode

A.W.L.Chiu L04 Page 7 of 8


BE520 – Brain Machine Interfaces

Bipolar
Adjacent electrode pairs
 Closely located electrode pairs can assist in measuring _______________________________
electric fields.
Common Average Reference (CAR)
Use the average of all active electrodes as reference

Equation:

CAR works best if Φ ( r R ) term goes to zero when _______________________ electrodes are used.

EXERCISE – Identify artifacts from EEG signals.

ELECTRODE TYPES
Passive vs. Active
Advantages of active electrodes:
 _____________________ for a broad range of scalp impedances.
 Tolerance against ______________________________________ across the scalp.
 Enhance _________________________ along the cable path toward the amplifier.

Wet vs. Dry


Advantages of dry electrodes:
 ___________________________________
Disadvantages of dry electrodes:
 ___________________________________

A.W.L.Chiu L04 Page 8 of 8

You might also like