Functional Magnetic Resonance imaging Flashcards
Name the key determinants of the fMRI signal from the observed stimulus to the observed activity
The stimulus is observed (or modulation in background activity), this causes neuronal activity. This results in neurovascular coupling. This haemodynamic response can then be detected by an MRI scanner which results in an fMRI BOLD response.
What is meant by a haemodynamic response and how does it cause the BOLD signal?
The BOLD signal stands for blood-oxygen level dependent signal. The haemoglobin in blood can bind with oxygen. Oxygenated haemoglobin (Hb) is diagmagnetic (zero magnetic moment). Deoxygenated haemoglobin is paramagnetic (magnetic moment). The BOLD Magnetic susceptibility if dHB is about 20% greater than Hb and this magnetic susceptibility affects rate of dephasing -T2 and T2* contrast.
What is this meant by this dephasing?
Tissues of blood vessels containing oxygenated haemoglobin are quite homogeneous. In comparison, theres an influence of deoxygenated haemoglobin inside the vessel on the tissue outside the vessel. After the Rf pulse the spins start to dephase back to B0. There is less dephasing in the oxygenated Hb than the deoxygenated Hb. This is because the dephasing is only determined by T2 in the oxygenated Hb, but i the deoxygenated Hb it is determined by T2*, so also by the susceptibility. This is reflected in the total magnetisation being reduced much faster in the oxygenated Hb than the deoxygenated Hb. The BOLD contrast is actually the difference in dephasing between the oxygenated and deoxygenated Hb.
Describe the BOLD response over time
After the stimulus onset there is an initial dip, an increase in the signal labelled the primary response, and then a negative overshoot.
If the BOLD signal increases with increasing neuronal activation is that not counter-intuitive?
This phenomenon was also puzzling to early fMRI investigators. You might imagine that if there is more activation, you will see neural activation increase and thus more deoxygenated Hb in that area. This decreases the signal because it decreases the difference between the oxygenated and deoxygenated Hb. This is what is observed in the initial dip.
What we now know happens, however, is that in response to brain activation, regional arteriolar dilatation with increased capillary filling occurs so more fresh (oxygenated) blood is supplied than is required for the brain’s immediate metabolic needs. This “overshoot” of oxygenated blood means that the relative concentration of deoxyhaemoglobin in activated areas will decrease. The T2/T2* shortening effects of deoxyhaemoglobin will be diminished and the BOLD signal in activated areas will therefore increase.
Describe the separate activity of the oxygenated haemoglobin and the deoxygenated haemoglobin over time after stimulus onset (SO)
The oxygenated haemoglobin started to go up 1-2 seconds after SO and peaks at around 4-5 seconds. The deoxygenated Hb first goes up a lil but then it then goes down as oxygenated Hb comes in and flushes out the deoxygenated haemoglobin.
When we consider the post stimulus undershoot, what parameters do we take into account? (3)
CBF; Cerebral blood flow: The amount of blood that flows into a region at a given time
CBV; Cerebral blood volume: The amount of blood in a certain volume
BOLD signal
Describe the activity of these three parameters during and after stimulus presentation (SP)
All of these parameters go up after stimulus presentation. The CBF and BOLD response goes down almost immediately after SP, however the CBV goes down much slower. This means that although there is not more blood flow coming in, the vessels remain larger. Thus the ratio between dHb and Hb goes down again. This is one theory for why we see this post stimulus undershoot.
How does this signal change during a prolonged BOLD response?
After the rise, following a small peak of the short peak is a longer ‘plateau’ before the subsequent fall and undershoot. This illustrates a prolonged increase of the BOLD response.
What are the “two main goals of fMRI task design”? What kind of issues are present with these two goals (broadly)
Create a desired cognitive state; Standard experimental design issues
Detect brain signals associated with that state; fMRI-specific experimental design issues
Ideally you could just have someone do a well designed task and the relevant brain area will light up, why is this not the case?
There is too much noise. You cannot simply look and identify the relevant activation from the ‘brain noise’ from activation irrelevant to the target construct or scanner noise resulting from the hardware.
How do you see through this noise (2)
You have to create a contrast; We subtract the activation of task A from control task B. While a single image doesn’t show much relevant brain activation, with contrast we are left only with the relevant signal.
We also need repetition; we need to repeat this stimulus because even though we do this contrast the signal is quite low. We need to do multiple averages to create sufficient signal to noise.
Name four ways in which we achieve the right contrast
Subtraction
Factorial
Parametric
Conjunction
Name two assumptions of the subtraction method
- Cognitive processes are additive
- Different cognitive processes don’t affect each other
To what extent are these subtraction methods valid?
These assumptions are often invalid but it does produce information that is practically useful
