Exam 3 Imaging MRI & fMRI Flashcards

1
Q

MRI

A

Magnetic Resonance Imaging

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

what is MRI? (aka, what deferentiates it from other imaging)

what does it do well?

A
  • •Imaging without exposure to ionizing radiation
  • •Offers better differential brain tissue contrast so better visualization of brain structures and differentiation of gray & white matter

Magnetic Resonance Imaging is a form of Imaging without exposure to ionizing radiation, It also offers better differential brain tissue contrast so better visualization of brain structures and differentiation of gray & white matter

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

in what planes are MRI performed?

A

MRI can be Done in horizontal, coronal and sagittal planes

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

How does MRI work?

A

•How it works

  1. –Oscillating magnetic field excites hydrogen atoms in the brain
  2. –Excited hydrogen atoms emitted radio frequency signals when they return to their equilibrium state
  3. –These signals are detected and converted into an image using algorithms
  4. –Contrast between different densities of neural tissue is determined by the rate at which excited atoms return to their equilibrium state
  5. –Image contrast may be weighted to demonstrate different anatomical structures or pathologies
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

how are contrasts between neural tissues work?

A
  • –Contrast between different densities of neural tissue is determined by the rate at which excited atoms return to their equilibrium state
  • –Image contrast may be weighted to demonstrate different anatomical structures or pathologies
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

what are four types of image weighting?

A
  1. T1
  2. T2
  3. PD (proton density)
  4. FLAIR
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What is T1 useful for?

A

•Image Weighting
–T1: Useful for visualizing normal anatomy Cerebrospinal fluid is dark

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

what is T2 useful for?

A

•Image Weighting
–T2: Useful for visualizing pathology. CSF is light, but white matter is darker than with T1

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

what is an easy way to remember T1 and T2 wieghting? (that dr. Wright told us)

A

Easy way to remember it is T2 = H2O for water

(water is light)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

what is PD (proton density) weighting useful for?

A

•Image Weighting
–PD (proton density): Better differentiation of gray & white matter with gray matter brighter than white matter & CSF appears brighter

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

what is FLAIR image weighting useful for?

A

•Image Weighting
–FLAIR: Useful in identifying demyelination & for evaluation of white matter plaques (b) compared to T2 (a)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

compare MRI vs CT (3 points each)

A

–CT

  1. •Less expensive
  2. •No magnetic precautions (implanted metal & devices)
  3. •Contrast media used can be toxic

–MRI

  1. •Better anatomical imaging
  2. •No radiation involved
  3. •Some have suggested to be safer in pregnancy
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is Diffusion MRI?

A

•Diffusion MRI

  • –Specialize form that measures diffusion of H2O molecules in biological tissues
  • –Since a H2O molecule inside a neuronal axon has a low probability of crossing the myelin membrane it will diffuse along the axon so axon will be highlighted by this technique
  • –Ideal for describing connectivity within the brain
  • –can determine the direction of water flow and this Direction can be determined easily with display color code for direction
    • This gives rise to illustrations that not only show connectivity but also show the direction of that connectivity which can be color coded.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

fMRI

A

Functional Magnetic Resonance Imaging

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

How does an fMRI work?

A
  • •Measures brain activity by detecting associated changes in blood flow
  • •Primary form uses Blood-oxygen-level dependent (BOLD) contrast
    • –Map neural activity by imaging the change in blood flow related to neuronal energy use
    • –Uses the change in magnetization between oxygen-rich and oxygen-poor blood as the basic measure
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Explain the physiology of an fMRI (blood flow), 5 things

A

•Physiology of fMRI

  1. –Increased blood flow from increased neural activity
  2. –Increased blood flow occurs within 2-3 mm of increased activity
  3. –Consumption of O2 decreases O2 saturation of blood in the area of activity
  4. –Deoxyhemoglobin has > magnetic iron leading to the > observed magnetic susceptibility effect
  5. –With > oxygenated blood arriving in the tissue a > contrast between oxygenated & deoxygenated blood so > contrast detected on fMRI

Increased blood flow results where there is increased neural activity, and the increases in blood flow occur within 2-3 mm of increased activity. What is measured is the Consumption of O2 which decreases O2 saturation of blood in the area of activity. And as previously mentioned there is a change in magnetization between oxygen-rich and oxygen-poor blood as the basic measure

This change in magnetization of blood comes from the fact that Deoxyhemoglobin has greater amounts of magnetic iron leading to the increase in the observed magnetic susceptibility effect of deoxygenated blood.

With a greater amount of oxygenated blood arriving in the tissue a greater contrast between oxygenated & deoxygenated blood so greater magnetic contrast is detected on fMRI

17
Q

Explain the physiology of an fMRI (voxels)

A
  • –Brain divided up into voxels (volume + pixel) of 1-25 cubic mm (50,000 neurons/cubic mm)
  • –Voxel size dependent upon
    • •Slice thickness
    • •Area of a slice
    • •Grid imposed on the slice by the scanning process
  • –Whole brain studies use larger voxel size so have millions of neurons and synapses

Now the Brain divided up into voxels the word voxel comes from the combination of volume + pixel and each voxel represents between 1-25 cubic mm depending upon the resolution of the fMRI. There are approximately 50,000 neurons/cubic mm. The Voxel size dependent upon

Slice thickness

Area of a slice

Grid imposed on the slice by the scanning process

Whole brain studies use larger voxel size so have millions of neurons and synapses in a single voxel

18
Q

where does the word voxel come from?

A

the word voxel comes from the combination of volume + pixel

19
Q

how big are voxels usually?

how many neurons are in each cubic mm?

A

–1-25 cubic mm (50,000 neurons/cubic mm)

20
Q

what are three things voxel size is dependent upon?

A

–Voxel size dependent upon

  1. •Slice thickness
  2. •Area of a slice
  3. •Grid imposed on the slice by the scanning process
21
Q

what is diffrent about whole brain studies in regard to voxels?

A

–Whole brain studies use larger voxel size so have millions of neurons and synapses

22
Q

Explain the physiology of an fMRI (contrast)

A

•Physiology of fMRI

  • –Contrast between voxels shown by gradation in depth of color
  • –False color to increase contrast

Now the Contrast between voxels shown by gradation in depth of color. This contrast can be enhanced using false colors.

23
Q

how is contraxt between voxels shown?

A

Now the Contrast between voxels shown by gradation in depth of color. This contrast can be enhanced using false colors.

24
Q

Explain the physiology of an fMRI (change in signal and response time)

A

•Physiology of fMRI

  1. –Change in the MR signal from neuronal activity is called the hemodynamic response (HDR)
  2. –Lag of 1-2 sec with lag of peak response 5 sec behind increase in neural activity
  3. –Neural responses are 10s-100s of msec
  4. –So 5,000 msec delay may not be the event but residual activity after the event – some have described this a “cellular memory” of the event

Change in the MR signal from neuronal activity is called the hemodynamic response (HDR). There is a Lag in increased activity of 1-2 sec with lag of peak response 5 sec behind increase in neural activity. Now Neural responses are 10s-100s of msec so you can see that a 5,000 msec delay may not be the event but residual activity after the event – some have described this a “cellular memory” of the event. Or it might be another event all together. So there is a temporal lag problem.

25
Q

the chane in the MR signal from neuronal activity is called ____________________.

A

–Change in the MR signal from neuronal activity is called the hemodynamic response (HDR)

26
Q

HDR

A

Hemodynamic Response

27
Q

what is the lag in increased acitvity?

how far behind is peak response?

A

There is a Lag in increased activity of 1-2 sec with lag of peak response 5 sec behind increase in neural activity

28
Q

how long is a neuroal response?

why is this significant?

A

Now Neural responses are 10s-100s of msec so you can see that a 5,000 msec delay may not be the event but residual activity after the event.

Or it might be another event all together. So there is a temporal lag problem.

29
Q

what is “cellular memory” of the event?

A

Now Neural responses are 10s-100s of msec so you can see that a 5,000 msec delay may not be the event but residual activity after the event – some have described this a “cellular memory” of the event.

30
Q

what are some additional issues with fMRI (besides the tempral probelm)?

A

•Some Issues about fMRIs

  1. –Intersubject variability – illustrated 6 subjects doing the same spatial memory task
  2. –Test-retest correlations comparing 6 different measures of same task ranged from 0.00-0.75 with a mean of 0.45
  3. –Repetition of same cognitive function shows up in different brain regions
  4. –Single brain regions show up doing widely different functions
  5. –Only show correlations not causation
  6. –Tested activities are generally very simple, isolated tasks – far from normal complex behaviors
  7. –Measurements blind to connectivity within the brain
  8. –Conclusion well beyond data
  9. In addition companies are trying to cash in on bad science.

There are some other Issues about fMRIs. There is a wide Intersubject variability – illustrated here are 6 subjects doing the same spatial memory task. Test-retest correlations comparing 6 different measures of same task ranged from 0.00-0.75 with a mean of 0.45

And even with questions about fMRI research, conclusions made by the popular press go way beyond scientific evidence

31
Q

Example of Conclusion Well beyond data: Truth machine

A

For many, establishing guilt or innocense is fMRI’s holy grail.

Study: Temple University

Protocol: Six graduate studetns were asked to fire a gun loaded with blanks, then lie about theri actions. Five students who didn’t fire a gun were told to be truthful. Could fMRI scans reveal who was lying?

Results: Fourteen areas of the brain, including the anterior cingulate cortex (top yellow dot) and the hippocampus (bottom), were active when subjects liedl seven areas were active when subjects told the truth.

32
Q

Example of Conclusion Well beyond data: Big Love

A

Love might be nothing more than a chemical reaction.

Study: State University of New York, Stony Brookl Albert Einstein College of Medicine; Rutgers University

Protocol: Researchers asked 17 young men and women to look at photos of the people they professed to love, then analyzed their brain activity in an fMRI scanner.

Results: Early stage romantic love is about motivation and reward, since it lights up subcortical reward regions like the right ventral tegmental area (top blue dot) and dorsal caudate area (bottom). Subjects in more extended romantic love showed more activity in the ventral pallidum (middle), which indicates attachement, in prairie voles - and, scientists surmise, in humans.

33
Q

Example of Conclusion Well beyond data: Proof of Purchase

A

One controversial use of fMRI is neuroeconomics - the study of mental and neural processes that drive economic decisions.

Study: Carnegie Mellon University, Stanford University, MIT Sloan School of Managment

Protocol: Twenty-six adults were given $20 each to spend on consumer items. Could researchers predict intent to purchase based on rain regions registering activity?

Results: When areas of the brain associated with product preference and evaluation of gains and losses - the nucleus accumbens (right red dot) and the medial prefrontal cortes (left), respectively - were activation, the person baught a product. Accuracy rate: 60 percent.

34
Q

Example of Conclusion Well beyond data: Better to Give

A

Does our brain think paying taxes is actually satisfying?

Study: University of Oregon

Protocol: Scientists gave 19 women $100 each, then scanned their brains as they watched their money go to a charity, via mandatory taxation and voluntary contribution.

Results: The caudate nucleus (right green dot) and nucleus accumbents (left), the same regions that fire when basic needs like hungar and social contact are met, were activated when subjects saw some of their tax money go to charity; activity was even greater when they gave money of their own accord. Scientists cite this as tentative proof of altruism.

35
Q

Example of Conclusion Well beyond data: The Oops Factor

A

What happens when you make a costly mistake?

Study: University of Michigan

Protocol: Scientists asked 12 adults to complete 360 visually based tests that carried monetary rewards and penalties between 25 cents and $2.

Results: When subjects made errors with consequences - in this case, losing money - the rostral anterior cingulate cortex (rACC, orange dot) was much more active. It was less active when mistakes carried no penalty. The rACC’s involvement suggests the importance of emotions in making decisions.

36
Q

what are new techniques combining fMRIs with?

A

New techniques are now Combining fMRIs with EEG recordings

37
Q

why are new techniques now combining fMRI with EEG recordings?

A
  • EEG temporal response instantaneous – msec delay – compared to over 10 sec with fMRI
  • EEG not spatially localizable – compared to the much better fMRI
  • Combine two techniques to enhance detection both temporally & spatially
38
Q

is there specially designed equipment to perform both fMRI and EEG recordings simultaniously?

A

yes

There is now Specially designed equipment to perform both fMRI & EEG recordng simultaneously