Introduction to EEG
EN3572-Biomedical Signal Processing Prepared by: Dr. Anjula C. De Silva Email: [email protected] Room: 3205, 1st floor, Sumanadasa building Slideshows: www.projectcuris.com/teaching.html
1
Definition of the EEG
The EEG consist of excitatory and inhibitory postsynaptic potentials of pyramidal cells generated in the cerebral cortex of the brain
The human EEG shows activity over the range of 1 to 30 Hz with amplitudes ranging from 20 to 300 V
Cerebral cortex
The outermost sheet of neural tissue of the cerebrum of the brain Functions
Memory Attention Perceptual awareness Thought Language Consciousness
Consists of six horizontal layers 2-4 mm thick
3
Pyramidal cells
Why not action potentials?
Action potentials do not contribute to the EEG due to the following reasons:
Less in time duration (high frequency transient) Asynchronous
only a small percentage of neurons fire action potentials at any one instant
Filtered out by the capacitive lipid membrane, which serves as a low-pass filter
reduce the ability to be recorded from surface electrodes
Why postsynaptic potentials (PSP)?
Slow waves
Even though low in amplitude compared to the action potential, more likely to overlap at any one instant and produce a substantial amplitude
Not filtered by the lipid membrane
Low frequency
Ability of temporal and spatial summation
Low frequency signals correspond to spatially larger currents and can therefore propagate farther
Synaptic functions of neurons
Action potentials transmit information from one neuron to the other However at synapses, alterations to these action potentials occur (processing)
Blocked from transmitting Changed from a single pulse into repetitive pulses Integrated with impulses from other neurons to form complicated pulse trains
7
Physiologic anatomy of the synapse
10,000-200,000 minute synaptic knobs called presynaptic terminals lie on the surfaces of the dendrites and soma of a neuron These presynaptic terminals are the ends of nerve fibrils that originate from many other neurons 80-95 % of them on the dendrites and only 5-20% on the soma
The transmitter vesicles contain the transmitter substance that, when released into the synaptic cleft, either excites or inhibits the postsynaptic neuron The mitochondria provides adenosine triphosphate (ATP), which supplies the energy for synethesizing new transmitter substance
Resting membrane potential of the neuron
Resting membrane potential: -65mV Basis of excitatory and inhibitory functions
Decreasing the voltage to a less negative value makes the membrane more excitable Increasing the voltage to a more negative value makes the membrane less excitable (inhibitory)
9
Resting membrane potential of the neuron
Sodium concentration gradient
caused by a strong somal membrane sodium pump that continually pumps sodium out of the neuron
Potassium concentration gradient
Caused by the potassium pump that pumps potassium to the interior
Chlorine concentration gradient
Caused by the negative voltage that repels the negatively charged chloride ions Could be due to a pump (can explain with Nernst potential)
10
Excitatory postsynaptic potential (EPSP)
When the presynaptic terminal secrets an excitatory transmitter into the cleft
Increase the membrane permeability to Na+ ions Rapid influx of Na+ ions into the soma due to
Concentration gradient Electro negativity
Increases the membrane potential to -45mV EPSP = +20mV
11
Spatial summation and reaching the threshold
A single presynaptic terminal can never initiate an action potential
Single EPSP 0.5-1 mV Duration
1-2 ms rising time while the Na+ gates are open 15 ms decay time to leak excess Na + through the pump
The highly conductive soma summate all the neuronal excitations that occur in the soma and the dendrites When the total EPSP becomes large enough
Threshold for firing is reached An action potential will generate at the axon hillock
High concentration of voltage-gated sodium channels in the axon compared to the soma and dendrites
12
Inhibitory postsynaptic potential (IPSP)
Inhibitory synapses open
Chloride channels
Allow Cl- into the membrane
Potassium channels
Allow K+ out of the membrane Movement of Cl- and K+
increase the intracellular negativity
Hyperpolarization
reduces the membrane potential to -70mV IPSP = -5mV
13
Nernst potential
Explanation of EPSP and IPSP can be supported by Nernst potential The potential that exactly opposes movement of an ion is called the Nernst potential for that ion
where EMF is the Nernst potential in millivolts on the inside of the membrane. The potential will be -ve for positive ions and +ve for negative ions
Concentration Concentration Nernst inside (mEq/L) outside (mEq/L) potential (mV) 14 142 +65 120 4.5 -86 8 107 -70
Ion Na+ K+ Cl-
14
Signals related to presynaptic and postsynaptic activity
15
Current flow in a dendrite of a pyramidal cell
Dendrites of pyramidal cells, which are oriented perpendicular to the cell surface, receive a variety of synaptic inputs Consider the flow of current produced by an EPSP on an apical dendrite of such a neuron Current flows (Na+) into the dendrite at the site of generation of the EPSP, creating a current sink It then must complete a loop by flowing down the dendrite and back out across the membrane at other sites, creating a current source
16
Dipoles
The polarity of the extracellular fluid forms a dipole
Current sink ve polarity Current source +ve polarity
The negativity at the apical dendrites near the surface is detected by EEG electrodes The potential recorded by the EEG is as a result of summed postsynaptic potentials from pyramidal cells that create dipoles between soma and apical dendrites
17
EPSP effect on the EEG electrode site
The situation is reversed if the site of the EPSP generation is on a proximal dendrite Polarity of the surface EEG depends on the location of the synaptic activity within the cortex
18
EPSP & IPSP effect on the EEG electrode site
The activity measured at a surface EEG electrode will have opposite polarities for these same two EPSP inputs, even though the basic electrical event is the same. EPSPs in superficial layers and IPSPs in deeper layers appear as upward (negative) potentials Whereas EPSPs in deeper layers and IPSPs in superficial layers have downward (positive) potentials Thus cortical synaptic events cannot be unambiguously determined from EEG recordings alone
19
Summed effect on the EEG electrode site
Single neuron activity is too small to be picked up by EEG EEG reflects the summation of the synchronous activity of many neurons with similar spatial orientations Each EEG electrode sees the summed activity of roughly 6 cm2 of underlying cortex
20
EEG recording
Effective bandwidth of EEG is 100 Hz
Typical sampling frequency is 200 Hz
Quantization 16 bits Electrodes
Disposable electrodes
Cannot use in places with hair
Reusable disk electrodes
stainless steal, tin, silver or gold plated
Needle electrodes Electrode caps
Impedance
Absolute electrode impedance < 5k Inter-electrode impedance < 1k
21
EEG electrode placement
International 10-20 system
International Federation of Societies for Electroencephalography and Clinical Neurophysiology
Developed to ensure standardized reproducibility
Over time in a patient/subject Among patients/subjects
"10" and "20" refer to the actual distances between adjacent electrodes are either 10% or 20% of the total frontback or rightleft distance of the skull
22
International 10-20 system
Anatomical landmarks used to position electrodes
Nasion: the depressed point where the top of the nose meets the ride of the forehead Inion: lowest point of the skull at the back of the head, normally felt as a prominent bump
Abbreviations
F: Frontal lobe T: Temporal lobe C: Central P: Parietal lobe O: Occipital lobe z: zero (midline) A: Earlobe Fp: Frontopolar Even numbers: represent electrodes on the right hemisphere Odd numbers: represent electrodes on the left hemisphere
23
Conventional electrode setting of 21 electrodes (excluding earlobes)
75 electrode setting based on American EEG Society
24
References
Guyton & Hall: Medical Physiology Andrew S. Blum & Seward B. Rutkove: The Clinical Neurophysiology Primer Sanei, SaeidChambers, J. A: EEG Signal Processing
25
