Image optimisation Flashcards

You may prefer our related Brainscape-certified flashcards:
1
Q

When is an MRI image optimal?

A
  1. High signal
  2. Low noise
  3. High tissue contrast
  4. Good spatial resolution
  5. Good temporal resolution
  6. Short imaging time
  7. No (or few) artefacts
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What are the basic image parameters?

A
  1. Sequence type:
    - 2D
    - 3D
    - TR
    - TE
    - Flip angle, TI
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is the flip angle?

A

The amount of rotation that the net magnetisation experiences during application of RF pulse

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

What is the resultant flip angle approximately proportional to?

A

Frequency of the B1 field

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

What are other image parameters?

A
  1. Number of averages, FOV, matrix size, bandwidth

2. Number of shots, slice thickness, echo train length

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

What is a measure of true signal to noise?

A

signal to noise ratio

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What does a lower signal-to-noise ratio generally result in?

A

Grainy appearance to images

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

What is the equation of SNR?

A

s/n

signal level/ noise level

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

What is a measure used to determine image quality?

A

CNR

Contrast to noise ratio

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

What are the 2 equations for CNR?

A
  1. SA-SB/standard deviation

2. Stissue1-Stissue2/n

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

What are MR images?

A

Complex valued (real and imaginary channels)

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

What is phase useful in?

A

Susceptibility mapping

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is Quantiative susceptibility mapping (QSM)?

A

Absolute concentration of iron, calcium and other subtanced may be measured in tissue based on changes in local susceptibility

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What do we refer to for noise?

A

Thermal noise

i.e. stochastic thermal fluctuation of voltage induced in the receiver coils

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

What does the thermal fluctuations of voltage include?

A

Contributions from:

  1. Imaged object
  2. Coil
  3. Electronics
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is noise in each image channel?

A
  1. zero-mean

2. Guassian-distributed

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What do magnitude images have?

A

More complex noise distributions such as Rician distributions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What are Rician distributions?

A

Models the path the scattered signals take to receiver

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What is the SNR calculations?

A
  • S: use mean signal in a homogenous region (e.g. within white matter)
  • N: use mean signal in an area with air
  • N: use standard deviation of signal in an area with air
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What is noise-only acquisition calculation

A
  • S: use mean signal in a homogenous region (e.g. within white matter)
  • N: use mean signal from noise-only acquisition (e.g. within same white matter region)
  • S: use mean signal in a homogenous region-of-interest (e.g. within white matter)
  • N: use standard deviation of signal from noise-only acquisition (e.g. within same white matter region)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What is SNR calculation: method 5?

A
  • S: use mean signal over the acquisition in each voxel

* N: use standard deviation of signal over acquisition in each voxel

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Why should we watch out for spatially-variant noise?

A

The noise level can vary across the image

- Watch out for image ghosts in ‘‘air’’ areas

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Why is the acquisition of multiple images be required as part of a quantitative study?

A

These images can then be exploited to estimate SNR

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What is time-consuming?

A

Acquistion of noise-only images

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

What does Contrast,C, measure?

A

Differences between the intensity of signal from two different tissues
- Differences is usually normalised

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

How can the contrast be changed?

A

Varying TR, TE, FA

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

How can the noise level,n, be estimated?

A

Studying areas with no tissue (i.e. air)

28
Q

What does CNR inform?

A

How relevant the contrast is with respect to random variations due to noise

-Aim: CNR > 1

29
Q

How is the SNR/CNR altered?

A
  1. Voxel size
  2. Number of averages (NEX)
  3. Number of frequencies encode steps
  4. Number of phase encode steps
  5. Bandwidth
30
Q

How can the total scan time be altered?

A

Changing parameters such as number of slices, number of shots, echo train length, acceleration factors

31
Q

How can the resolution be altered?

A

The volume of the voxel

V= ∆x ∆y ∆z

32
Q

What is bandwidth?

A

The inverse of dwell time

33
Q

What is the dwell time?

A

Sampling time along the frequency encoding direction

i.e. time separating two points when filling k-space

34
Q

What are examples of acceleration of acquistion?

A
  1. Partial Fourier Imaging
  2. Parallel Imaging
  3. Simultaneous multi-slice imaging
35
Q

What is Partial Fourier Imaging?

A
  1. Only a fraction of k-space is sampled

2. It relied on symmetry properties of Fourier Transform

36
Q

What is parallel imaging?

A
  1. The K space is sampled below the Nyquist limit
37
Q

What is simultaneous multi-slice imaging?

A
  1. Two or more slices are excited together using specialised RF pulses
  2. The sensitivity of multiple coils is exploited to unfold slices
38
Q

What implies trading off?

A

Designing an MRI sequence for practical use

39
Q

What are the steps for trading off?

A
  1. Select contrast required for application
  2. Set up FOV and resolution
  3. Adjust other sequence parameters taking time into account
40
Q

What is an artefact?

A

Feature that appears in MRI image but that is not present in image’s tissues

41
Q

What is imaging artefact essential to?

A

Acquisition technique and how it interacts with underlying physiological processes

42
Q

What may artefacts be misinterpreted for?

A

pathology

43
Q

What are artefacts?

A

often systematic and in a way reproducible

44
Q

What differs from thermal noise?

A

Imaging artefact

45
Q

How to minimise motion artefact?

A

Use of fast imaging sequences

46
Q

What happens when you move during an acquistion?

A
  • Tissues moves from one place to another – has the ‘wrong’ frequency and phase
  • This can lead to signal losses (imperfect rephasing) or signal increases
47
Q

What is physiological noise?

A

• Related to motion due to normal physiological processes
- Pulsation of cerebrospinal fluid
- Respiration
• Lead to signal variability over time – have physiological origin
• Can be mitigated by synchronizing the acquisition with cardiac or respiratory cycle (gating)
- Variable TR (i.e. variable T1-weighting)
- The acquisition is usually less efficient and can last longer

48
Q

What is spike noise?

A
  • Artefact caused by loose cable in gradient system
  • Causes noise spike in k-space leading to oscillatory patterns in image space
  • Call engineer
49
Q

What is zipper artefact?

A
  • Artefact caused by leakage of RF radiation at a specific frequency
  • Can be internal – scanner fault
  • Or external – leakage through shielding
50
Q

What are tissues characterised by?

A

Different magnetic susceptibility

51
Q

What does magnetic susceptibility describe?

A

How a tissue interacts with the applied field

52
Q

Where do strong gradients arise?

A

At boundaries of tissues with different magnetic susceptibility

  • Rapid dephasing of spins - signal loss
  • This changes the effective gradient - distortion, signal pile up/drop out
53
Q

What is highly prone to susceptibility artefact?

A

Echo planar Imaging

54
Q

What are magnetic susceptibility artefacct?

A

Refer to a variety of MRI artifacts that share distortions or local signal change due to local magnetic field inhomogeneities from a variety of compounds

55
Q

How to mitigate susceptibility artefact?

A

• Reduced EPI echo train length
• Use multiple EPI shots
• Correct image after acquisition with a B0 field map
- Estimated studying the phase at different echo times
- Estimated from 2 magnitude images with reversed phase encode direction

56
Q

What are Eddy Current Artefact?

A

Current loops induced in the metallic part of the scanner by fast switching gradients

57
Q

What do Eddy Current Artefact cause?

A

Additional, undesired magnetic field that lead to distortion or imperfect rephasing

58
Q

Where is Eddy Current Artefact an issue in?

A

Diffusion MRI

Mitigated with post-processing techniques

59
Q

How to mitigate chemical shift artefact?

A
  1. Increase the bandwidth

2. Use dedicated fat suppression techniqye

60
Q

What are the Wrap Around Artefact?

A

When the object producing MR signal is larger than the FOV or when the FOV is misplaced

61
Q

What are the Gibbs Ringing or Truncation Artefacts?

A
  1. Commonly seen artefact
  2. Bright and dark bands at high contrast boundaries
  3. Caused by finite k-space sampling
62
Q

How is the Gibs Ringing or Truncation Artefact reduced?

A

Increasing the matrix size or with post-processing techniques

  • De-ringing filter
  • K-space extrapolation
63
Q

What are the other artefacts that exist?

A
  1. Gradient non-linearity

2. Phase-cancellation

64
Q

What is gradient non-linearity?

A

When the gradients are non-linear

Additional image distortions are seen

65
Q

What are phase-cancellation?

A

When the fat and water signal are 180 degree out of phase,

they cancel each out leading to signal loss