3 Magnetoencephalography (MEG) Flashcards
- What does MEG measure?
Magnetoencephalograpy (MEG) records magnetic fields produced by electrical currents occurring naturally in the brain, using very sensitive magnetometers.
- MEG vs EEG list three charactestics for each.
EEG: Electrocal potential (voltatge V)
- difference between two places
- Reference electrodes (earlobes, nose etc)
- Electrical circuit
MEG: Magnetic field (T)
- can be measured at a single point
- no need to referece signal
- no electrical circuit
- EEG vs MEG
Which has the better spatial resolution?
EEG electrondes complete a circuit with impedence from skin, scull etc which reduces the measurment voltage
In contrast with MEG the signal is unimpeded by skull etc and the siganal is less spread out creating better resolution.
- Does MEG have good spatial resolution?
MEG has relatively low spatial resolution
However -
MEG does have very good temporal resolution ms by ms
- To generate a signal that is detectable, approximately ……………. active neurons are needed.
It is often the layer of ………………….., which are situated perpendicular to the cortical surface, that gives rise to measurable ………………………
To generate a signal that is detectable, approximately 50,000 active neurons are needed.
It is often the layer of pyramidal neurons, which are situated perpendicular to the cortical surface, that gives rise to measurable magnetic fields.
- What is a dipole?
A dipole can be described as a magnet with a positive and negative end.
Technically: A pair of equal and oppositely charged or magnetized poles separated by a distance.
A dipole can be used to describe activity in several parts of the cortex.
- MEG detects ………………. currents associated primarily with ………………. …………………..
MEG detects intracellular currents associated primarily with postsynaptic potentials.
- What is the right hand rule? Radial or tangential dipoles?
According to the right-hand rule, a current dipole gives rise to a magnetic field that flows around the axis of its vector component.
- Radiale dipole can’t be picked up by MEG
- Tangential dipole are in the sulci
- Does MEG measure the gyrus well?
No it measures the sulci (a depression or groove in the cerebral cortex)
- Will a tangential source give different readings for MEG vs EEG?
Yes - they have different sensitivities.
- Is it hard to pick up sulci readings?
Yes - radial readings are “easier” as they are on the surface for EEG. MEG can only pick up sulci though but at a far more efficient rate than EEG can.
As an example:
If a study hypothesises that there is increased brain acticity within the sulci then perhaps an MEG is worth persuing (though expense is a big factor!)
- What is one of the biggests challanges for picking up the magnetic potentials?
There are magnetic fields all round us naturally - the earth itself has them and as the human brain has very small magnetic fields MEG.
MEG must boost the signals and minimise interference from external sources.
- How is noise minimised for MEG?
The MEG scanner is placed in a faraday cage (a metal box sheilded from outside elctrical and magnetic influences)
An axial gradiometer has a componsatory coil (secondary to coil to measure brain) to remove background noise.
- What is a SQUID?
Superconducting Quantum Interference Device
Liquid Helium needs to replaced at certain intervals.
Hi tech and expensive > 1million to buy!
100k per year to run (needs helium - engineers and support staff)
In comparison EEG is very cheap!
- Present-day MEG arrays are set in helmet-shaped ……… that typically contain ……….. sensors, covering most of the head. In this way, MEGs of a subject or patient can now be accumulated rapidly and efficiently.
Present-day MEG arrays are set in helmet-shaped dewar that typically contain 300 (more than 275 as stated in lecture) sensors, covering most of the head. In this way, MEGs of a subject or patient can now be accumulated rapidly and efficiently.
- How is more noise removed from the MEG data?
- Reference is subtracted from the initical SQUID output
- Filters remove the unwanted signals (Lowpass, hipass 50-100Hz filter)
- Averaging final output (ideally 100+ trials)
- Most MEG machines are ……………….
The machine is very sensitive so any ……….. items are not permitted in the scanner.
Most MEG machines are upright.
The machine is very sensitive so any metal items are not permitted in the scanner.
- Where are the sensors placed?
Problem: If the head moves the sensors position over the brain moves
Solution: Makers are placed on the scull at certain points - the nasia Fiducial points.
19.
Give an example of a study using MEG
A visual tactile cueing experiment.
- Pay attention to LEFT vs RIGHT
- Respond to TOUCH or VISION
Cued vs Uncued - the brain has a stronger response if the stimuli is cued.
The somoatosensory cortex - folded into the central sulcus was a good use of MEG.
- What is ERF?
ERF stands for Event Related Field and is similar to ERP - event replated potemtial.
A typical set of trials will give a relatively stable pattern for each ERF. These may look similar to those ERPs found in EEG.
Negative is BLUE positive is RED
- The siganls from MEG give much more than ERFs - name another set of data.
Time-frequency decomposition.
Using fourier transforms the power of frequencies can be plotted.
- Can time-frequency decomposition be visualised on head images?
Yes the data is mapped into the frequency spectrum and can be displayed showing the areas of activation on images.
24.
What is the problem with source reconstruction?
Activity is measured and it has a particular strength - we need to find out where that activity came from.
A given meaurement could arise from an infinite number of possible dipoles.
25.
How can the correct dipole be selected from MEG data?
There is no correct answer - We can only see which one fits better.
All available information is put into a program to be analysed.
26.
What is the forward model?
A program is given data and generats lots of possible models.
Using the data we can find the best possible solution (less error). Choose one that fits the data best (this could be many).
27.
If there is less than ……% risidual varience after the ……………….. ………….. of a forward model most researchers are happy to take the solution as reliable.
If there is less than 5% risidual varience after the iterative process of a forward model most researchers are happy to take the solution as reliable.
28.
What is the potential problem with timing?
Once a stimuli is presented readings need to be taken very quickly (20ms - 60ms) ideally otherwise secondary brain areas may become active in responce to the initial activation (by 100-200ms large brain areas may be active and no data at this point would be very useful for most research).
The data is still descriptive - never proof.
32.
Describe MEG response from string players Elbert et al (1995).
String players showed a larger response to controls in finger sensitivity of the left hand.
The arrows represent the location and orientation of the ECD (equivalent current dipoles) vector for each of the two digits’ averaged across musicians (black) and controls (yellow). The length of the arrows represents the mean magnitude of the dipole moment for the two digits in each group. The average locations of D5 and Dl are shifted medially for the string players compared to controls; the shift is larger for D5 than for Dl. The dipole moment is also larger for the musicians’ D5, as indicated by the greater magnitude of the black arrow
MEG was good as it gave a clean set of data from the central sulcus.
34.
Kirchner et al (2004). MMN Responses in Schizophrenia ?
What happens in the planum temporale?
A combined fMRI and MEG study.
MMN (miss matched negativity) - presenting participant with train of auditory stimuli. At certain intervals a sound will be heard that is not attended to. Occasionaly there was a deviant pitch.
The brain will pick up that there is a change in stimuli (a negativity is found at around 250ms)
The planum temporale may be responsible - MEG was good for working with the gyrus and ignoring the radial data.
35.
Kirchner et al (2004). MMN Responses in Schizophrenia ?
A combined fMRI and MEG study.
What difference against controls?
Waves show that even when not attending to audio stimuli the brain will note differences.
Schizophrenics have reduced MMN (miss matched negativity)
