Objectives

  • 2.1.34 Electroencephalography (EEG)
    • 2.1.34.1 Physiological basis of normal EEG and common EEG abnormalities
    • 2.1.34.2 Recognition of normal physiological rhythms in wakefulness, drowsiness and sleep
    • 2.1.34.3 Principal characteristics of neurophysiological maturation in children
    • 2.1.34.4 EEG indications and limitations, including sleep-deprived, video, intensive care monitoring and ambulatory EEG
    • 2.1.34.5 Recognition of common EEG abnormalities and their significance
  • What does the scalp EEG record?
    • summated electrical changes of the underlying cortex
    • occasionally more distant brain
    • extracerebral activity (artifacts)
  • signal amplitude depends on
    • intensity of source signal
    • distance from recording electrode
    • spatial orientation of generator
    • electrical resistance and capacitance of circuit between source and electrode
  • It is generally assumed that cortical (esp. pyramidal) neurons are the principal generators of the EEG
    • an individual neuron generating an electrical potential acts as a dipole
  • Potential changes will be best recorded when they are
    • occurring near the recording electrode
    • generated by cortical dipole layers that are oriented toward the recording electrode (perpendicular to the scalp)
    • generated in a large volume of tissue
      • comparison of simultaneous ECoG and scalp EEG suggests that 6 cm2 of cortex with synchronous activity is needed to generate a signal that is 'seen' on the scalp
    • rise and fall at rapid speed
  • scalp EEG only rarely records activity at distant sites

Electrode Placement

  • According to the international 10-20 system
  • important landmarks
    • inion
    • nasion
    • preauricular point

Differential Amplifiers

  • are used to address the problem of "common mode" signal among EEG channels
    • there is no "neutral reference" in EEG - everything is measured relative to something else
  • amplifiers increase the output signal amplitude so that even small potentials are seen
    • but this amplifies undesired signal as well, ie noise
  • the solution is to use differential amplifiers that amplify only the difference between input signal and reference point, by common mode rejection
  • the quality/amount of discrimination is often expressed as a ratio of the differential and common mode amplification of an amplifier
    • this is called the common mode rejection ratio (CMRR)
    • CMRR is the ratio of the common mode input voltage over the output voltage
    • in other words, it is how "loud" the common mode signal has to be before it "drowns out" the differential input signal (and in fact CMRR can be expressed in dB)
    • amplifiers with high CMRR are better at separating noise from desired signal
  • failure of discrimination to eliminate common artifact is usually due to
    • unequal impedance of the recording electrodes
    • absence of an effective ground connection to the patient
(examples of a differential amplifier in action)

Filtering the EEG signal

  • Types of filters
    • low frequency filter (high pass filter) - removes low frequencies, allows high frequencies to pass
    • high frequency filter (low pass filter) - removes high frequencies, allows low frequencies to pass
    • notch filter - to remove a specific frequency (ie 60 Hz)
  • Capacitors are the basis of filters
    • they are ideal for filtering alternating signals such as EEG recordings
    • they offer very high impedance (resistance to alternating current) to low/zero (DC) signal, but their impedance decreases as signal frequency increases
    • a capacitor in the direct path of current flow will filter out lower frequencies creating a high-pass filter (or LFF)
    • a capacitor placed between the current path and ground will "siphon off" higher frequencies (by offering a lower impedance route), leaving slower frequencies to pass by, acting as a low-pass filter (or HFF)
    • (model filter circuits)
    • Filters are not absolute - they do not act like "brick walls"
      • filters gradually eliminate frequencies above and below the cutoff frequency of the filter
      • filters differ in their frequency response, that is the degree to which they attenuate a given range of frequencies
      • (examples of filtering a model alternating signal)

Source Localization

  • Uses the EEG to try to solve the inverse problem
    • that is, describe the set of electric potentials that explains the recorded waveforms
    • theoretically impossible as an infinite number of combinations of dipoles can produce a given set of recorded potentials
    • in practice a reasonable guess can be made in many cases by using our knowledge of brain anatomy and physiology to limit the number of possibilities
  • Spatial localization is more important than pattern recognition
    • the EEG is nothing more than the difference in voltage between different electrode inputs expressed over time
    • the absolute potential at any individual electrode can never be known
    • there is no "reference-free" information, no "ideal zero" to compare to
    • this makes it necessary to use multiple combinations (montages) of different electrode recordings to arrive at an estimate of activity at a given location

What is a montage?

  • A montage is a combinations of electrode derivations
  • Montages perform the function of spatial filtering
    • they filter out similarly shaped waveforms that are simultaneously and widely distributed over the scalp
  • 5 basic kinds of montages
    1. bipolar montage
  • reference montages
    1. common electrode reference
    2. average reference
    3. weighted average reference
    4. Laplacian (source derivation)

Bipolar montages

  • Bipolar derivations are created by subtracting the potential of neighbouring electrodes
  • Typically bipolar derivations are linked in straight lines, or "chains"
    • commonly used "double banana" montage
  • Strengths
    • bipolar montages act as spatial filters that remove similar amplitudes and phases (coherent waveforms) from the recording
    • well-suited for analyzing low to medium amplitude waveforms that are highly localized
    • phase reversal on a bipolar montage may have localizing value (see figure)
  • Caveats
    • a waveform that points up is no more positive than a waveform that points down
    • polarity depends on the input of the differential amplifier the active electrodes connects to
    • ***A phase reversal is not always an abnormal or pathological finding - they are simply the byproduct of the configuration of the bipolar montage

Reference montages

  • Reference montages are all alike in that each input of the EEG is compared to another input that is the same (or at least similar) across derivations
  • the various methods are trying to make a "better reference"
    • common reference - each electrode in input 1 is compared to the same electrode for input 2 in all channels (e.g. Fz or A1/A2)
    • average reference - each electrode in input 1 is compared to the average of multiple channels for input 2
  • signals recorded on reference montages tend to be higher amplitude than bipolar due to longer inter-electrode distances
  • comparison of potential amplitude can be useful for source localization
  • Caveats
    • doesn't filter out widespread potentials with similar amplitudes and phases (i.e. coherent waveforms), so smaller, more local phenomena can be missed
    • "reference contamination" issues - a potential that affects only the reference electrode may influence the output in all channels, especially if it is large amplitude
  • schema of referential montage recordings

An Orderly Approach to the EEG

  • Two pieces of clinical information are necessary:
    1. patient age
    2. level of alertness
    • medications, co-morbidities, and reason for test are helpful
  • Comment on the technique and montage used:
    • state whether it is a routine, sleep-deprived, or telemetry EEG (surface or depth electrode)
    • mention sources of extracerebral artifact if they significantly obscure the recording
  • Describe the findings in an orderly, systematic fashion: (descriptors)
    1. Epileptiform abnormalities
      • waveform morphology (spike, sharp wave, spike & slow wave)
      • location (focal, lateralized, or wide-spread)
      • frequency of occurrence, or evolution during the recording
    2. Background rhythm
      • frequency bands (δ <4Hz; θ 4-7Hz; α 8-13Hz; β >13Hz)
      • location (focal, lateralized, or wide-spread)
      • persistence during the recording
    3. Response to activating procedures (hyperventilation, intermittent photic stimulation)
    4. Sleep, if attained
    5. EKG
    6. Summary of findings and interpretation

Descriptors of EEG activity

  • Even if you don't know what it is, you should be able to describe it!
  • Any wave can be described in terms of:
    1. wave form (e.g. morphology) (examples)
      • monophasic, diphasic, triphasic, polyphasic
      • regular (sinusoidal, saw-toothed, arch-shaped), irregular
    2. repetition
      • rhythmical repetitive waves (sinusoidal, spindles)
      • arrythmical irregular waves
    3. frequency
      • δ <4Hz; θ 4-7Hz; α 8-13Hz; β >13Hz
    4. amplitude
      • low <20μV; medium 20-50μV; high >50μV
      • relative amplitude differences (e.g. asymmetries) are more important than absolute amplitude
    5. distribution
      • widespread, diffuse, or generalized (used interchangeably)
      • lateralized
      • focal
    6. phase relation
      • timing and polarity of waves between multiple channels
      • described in degrees
      • waves that are 180° out-of-phase are described as a "phase reversal"
    7. timing
      • simultaneous or synchronous activity in multiple channels
      • asynchronous activity occurs at the same time in multiple channels, but without constant time relation to each other
      • independent activity arises from different regions, at different times
    8. persistence
      • how often a wave or pattern occurs during the recording
      • may simply report an estimate of the proportion of time during which the activity appears (e.g., 50%)
      • infrequent activity may be called sporadic or intermittent
    9. reactivity
      • describes how the pattern changes in response to various maneuvers
      • pattern may be increased (induced), or decreased (blocked)
      • e.g. eye opening, eye closure, hyperventilation, IPS, changes in level of alertness, movements, attempts to alert patient

The alpha rhythm

  • frequency band of 8-12 Hz
    • slower frequencies in childhood
  • maximal amplitude in posterior head regions, but may be seen in wide-spread distribution
    • amplitude asymmetry is common, with higher amplitude over non-dominant hemisphere
  • present only during wakefulness, with the eyes closed (attenuates with eye opening)

The beta rhythm

  • frequencies >13 Hz
  • often maximal amplitude in fronto-central head regions, but may be seen in wide-spread distribution
  • persistence and incidence of beta rhythms rises with age
  • excessive prominent beta activity is often attributed the effects of medications (esp. sedatives)

Slow wave activity

  • Slow wave activity consists of delta and theta frequencies
    • recall δ <4Hz; θ 4-7Hz; α 8-13Hz; β >13Hz
  • It is a non-specific finding that may correlate with cerebral pathology
    • normally present in sleep, or during wakefulness in infants and elderly adults
  • Generalized slow wave activity is suggestive of widespread or diffuse cerebral pathology
    • e.g. toxic-metabolic encephalopathy, hypoxic-ischemic encephalopathy
  • Localized slowing is more suggestive of a focal cerebral lesion
    • e.g. the "Blevins tetrad" - trauma, tumor, vascular, inflammation
  • In general, slowing is more likely due to cerebral pathology if it consists of:
    • persistent rather than transient slowing
    • delta rather than theta activity

References

  • Fisch & Spehlmann's EEG Primer
  • Gloor, P. (1985). Neuronal generators and the problem of localization in electroencephalography: application of volume conductor theory to electroencephalography. Journal of clinical neurophysiology : official publication of the American Electroencephalographic Society, 2(4), 327-354.

Introduction to EEG

August 9th, 2013

printer-friendly
(-)