Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

*FRCR1 Physics > Mersey MRI > Flashcards

Mersey MRI Flashcards

1
Q

Magnetic Resonance (MR)
* Unpaired protons (or neutrons) in the atomic nucleus have
a net magnetisation. Unpaired protons can be found in
hydrogen nuclei.
* These magnetic properties can be stimulated to produce a
measurable signal, but only at a specific (resonant)
frequency

A
  • MR signals can be generated from cellular water and/or fat
    molecules
  • To generate an MR signal:
    – Place the patient in a strong magnetic field, Bo
    – Expose the patient to a burst of EM energy at the resonant
    or Larmor frequency (this is known as an RF pulse)
    – Measure the signal (voltage induced in a coil of wire)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

Magnetic Properties of Protons
* All protons are associated with a tiny magnetic field
* The magnetic moment,  determines the magnitude and direction
of this magnetic force
* In the nucleus protons are paired with their magnetic moments in
opposite directions – which cancels out the magnetisation.

A
  • Nuclei with unpaired protons will have a net magnetic moment
    which can be exploited to produce an MR signal/image
  • Hydrogen nuclei comprise single unpaired protons & are the source
    of MR signal in clinical imaging
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Magnetic Moments are Vectors….
The magnetic moment of a proton is a vector quantity. This means both
its magnitude & direction are important.
We can represent the magnetic moment as an arrow: the length of the
arrow is proportional to the magnitude, while the arrow head gives the
direction of the vector..

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

Combining Vectors

When vectors are combined both the
direction & magnitude must be considered.

In these examples the yellow vector is the
sum of the 2 blue vectors

A

In the same way, the magnetic
moments from a group of protons
can be combined into a single net
vector.

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

Placing protons (water/fat) in a magnetic field

A

Protons in a magnetic field

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

Placing protons (water/fat) in a magnetic field

A

Bulk Magnetisation Vector, Mo

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

RF Pulses

The strength & duration of the RF pulse
controls the flip angle. Any size of flip angle can be achieved.

A 90 degree pulse rotates Mo

through 90 degrees
into the transverse plane
A 180o pulse inverts Mo

A

Magnetisation Vectors

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

RF Pulses

  • Need to stimulate Mo (get it to absorb energy) to generate an MR signal
  • Electromagnetic (EM) radiation consists of rapidly oscillating electric & magnetic & fields
  • So, exposing the sample/patient to a short burst of EM energy should stimulate Mo
A
  • Mo only interacts with the EM radiation at 1 specific frequency: the Larmor frequency
  • The Larmor frequency is the precession frequency of the protons
  • For protons in a 1T field, Larmor freq = 42MHz, which is in the radio wave section of the EM spectrum
  • Hence the stimulating EM radiation is referred to as an RF (Radio Frequency) pulse
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

The MR Signal

A

MR Relaxation

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

T2 Relaxation
(Spin-spin relaxation)

After the RF pulse the protons in the
sample are spinning in phase (they are in
step with each other).

Spins quickly lose coherence and get out
of step with each other due to local
variations in magnetic field strength. Mxy
(ie the MR signal) gets smaller.

The rate at which the signal decreases
with time depends on the T2 of the tissue
T2 is the time taken for the signal to
reduce to 37% of its original value.

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

T1 Relaxation
(Spin-lattice relaxation)

A 90o RF pulse causes some
protons to flip from “spin up” to
“spin down”. Afterwards these
protons slowly flip back causing
Mz to get bigger.

The rate at which Mz recovers is governed by the T1 relaxation
time of the tissue.

T1 is the time taken for Mz
to reach 63% of its full value.

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

Larmor Frequency

  • The Larmor Frequency is the resonant frequency at which energy is
    absorbed by protons & also the frequency of the MR signal.
  • Larmor Freq = γB, where
    – γ is the gyromagnetic ratio for the nucleus

– B is the net magnetic field experienced by the nuclei

A
  • The net magnetic field experienced by protons in the body
    depends on:

Includes susceptibility effects at tissue boundaries

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

Dipole-Dipole Interactions

Magnetic dipole of a proton
extends into space around it

In a water molecule the magnetic dipoles of the protons can interact with each other causing a local variation in magnetic field strength and therefore a change in the Larmor frequency.

This is an example of an intramolecular dipole-dipole interaction.

A

Unpaired electrons and/or protons in other molecules can interact with the proton dipoles in the molecules generating the MR signal.

Due to molecular motion and rotation these interactions are random (unpredictable). The rate of molecular motion affects MR relaxation
times. Large molecules produce slowly varying magnetic fluctuations at the molecular level.

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

Dipole-Dipole Interactions

The rapid movement of small molecules (ie unbound water molecules) creates rapid fluctuations in the magnetic environment at the molecular level. These fluctuations are too rapid to cause T2 relaxation (dephasing of the MR signal) and too rapid to stimulate protons to flip between energy states (T1 relaxation)

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

T2 Relaxation (1)

Immediately after a 90o RF pulse the protons precess in phase to
produce the MR signal
* Dipole-dipole interactions cause local variations in magnetic field
strength which quickly causes different precessional frequencies
across the spin population (remember, the Larmor frequency is
proportional to the net magnetic field at any point)

A
  • With the protons spinning at different rates they quickly get out
    of step with each other (ie they lose phase coherence)
  • Consequently, the MR signal diminishes
  • Phase coherence is lost most readily if the surrounding molecular
    motions or vibrations are relatively slow
  • Phase coherence is also affected by inhomogeneitiesin the main
    magnetic field strength, Bo
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

T2 Relaxation (2)

  • T2 relaxation occurs due to protons precessing at different rates
    resulting in a gradual loss of phase coherence. Hence it is also known as
    “spin-spin relaxation”
  • Molecular motion in water (& other small molecules) is relatively fast
    and creates rapidly changing local magnetic fields
  • Hence, water & CSF have relatively long T2 relaxation times
A
  • Protons in fat molecules, or those that are bound to protein molecules
    will experience slower molecular vibrations
  • Hence, fat has a relatively short T2 relaxation time
  • Spin-spin relaxation due to molecular motion is characterised by the
    T2 relaxation time
  • Spin-spin relaxation due to molecular motion & non-uniformities in Bo
    is characterised by the T2* relaxation time
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

T2 Relaxation Times

T2 relaxation times govern how quickly the MR signal diminishes from a particular tissue.

Because T2 relaxation varies with tissue type, the MR signal will decay at
different rates – this can be exploited to produce image contrast.

A

T2 vs T2*

The loss in MR signal with time is due to loss of phase coherence caused by:
* Magnetic interactions at the molecular level (spin-spin relaxation)
* Static differences in the magnetic field that remain constant over time within
a specific location (caused by the patient’s body as well as inhomogeneities of
the main magnet)
* T2 represents signal loss due to spin-spin relaxation only
* T2* is a combination of spin-spin relaxation & magnetic field inhomogeneity

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

T1 Relaxation (1)

  • During an RF pulse the spin (proton) population absorbs energy
  • The spins must lose this energy to return to their equilibrium state
  • Protons in tissue are exposed to constantly varying local magnetic
    fields due to random molecular motion.
  • These fluctuations are caused by the rapid movement of magnetic
    dipole fields from unpaired protons in the nuclei of neighbouring
    molecules
A
  • These tiny magnetic variations are superimposed on top of the main
    external field, Bo
  • Those molecular vibrations that occur at the Larmor Frequency will
    act like mini RF pulses and will stimulate the protons to lose energy
    to their surroundings.
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

T1 Relaxation (2)

A

T1 Relaxation Times

Because T1 relaxation varies with tissue type, Mz will recover at
different rates – this can be exploited to produce image
contrast.

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

MR Relaxation Times

Spin-lattice (T1) relaxation requires molecular motion/vibration close to the Larmor
frequency, which changes with field strength. It follows that T1 relaxation time for
tissues will also depend on Bo

. T1 increases with Bo
T2 relaxation requires slowly varying molecular magnetic interactions regardless of Bo

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

T1 weighted Images

A short TR prevents full recovery of Mz
between repeated pulses and generates
MR signals that depend on the T1
relaxation times of the tissues.

A short TE prevents the signals
decreasing as a result of T2
relaxation before the echo
signal is readout.

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

The MR Pulse Sequence

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

The MR Pulse Sequence

A
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Parameters for Controlling Image Contrast

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

T2 weighted Images

A long TR allows full recovery of
Mz between repeated pulses
and generates MR signals that
are independent of each tissue’s
T1 relaxation

A long TE allows the MR
signals to diminish according
to the T2 relaxation time of
each tissue before the echo signal is readout.

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

Proton Density Image

A long TR allows full recovery of
Mz between repeated pulses
and generates MR signals that
are independent of each tissue’s
T1 relaxation time.

A short TE prevents the
signals decreasing as a result
of T2 relaxation before the
echo signal is readout.

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

Image Contrast in T1 & T2 weighted images

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

MR Signal Size

The MR signal from any voxel is
proportional to:
* The proton density in the voxel
* The voxel size

A
26
Q

MR Signal Size

A
27
Q

Noise in MR

Noise in MR images is generated from the whole patient through Brownian
motion of biological molecules. This generates weak, rapidly fluctuating
magnetic background signals which are detected by the receiver coils.

A

These signals are emitted across a wide range of frequencies (ie ‘white noise’).

The receiver coils themselves also generate electronic noise, which is added to the noisy signals from the patient.

28
Q

Signal-to-Noise Ratio (SNR) in MR

A
29
Q

Spatial Resolution in MR

  • Spatial resolution in MR is governed by the size of the tissue
    volume (voxel) that generates the MR signal
  • Smaller voxels means the signal from the patient is sampled at
    smaller intervals allowing visualisation of finer detail better, ie better spatial resolution)
A
  • Note – smaller voxels will generate smaller signals so SNR will
    decrease
  • There is a trade off between spatial resolution and SNR
  • Voxel size = pixel size x slice thickness
  • Pixel size = Field of view / Matrix size
  • Matrix size = number of pixels along axis of image
  • Field of View = area of patient to be imaged (mm x mm
30
Q

Spatial Resolution in MR

A
31
Q

MR Signal Localisation

  • To produce an MR image, we need to map signals generated from individual voxels of tissue (this is signal
    localisation).
A
  • Must apply 3 magnetic gradient fields
    – Slice select gradient
    – Phase encode gradient
    – Frequency encode (or “readout”) gradient
  • The gradients are switched on and off repeatedly during any pulse sequence
32
Q

Magnetic Field Gradients

Change the value of the Larmor Frequency ALONG that axis

A

Slice Select Gradient (SSG)

33
Q

Slice Thickness

Larger if
- WIDER BANDWIDTH
- BROADER GRADIENT

A
34
Q

Phase Encode Gradient (PEG)

  • The PEG is briefly applied in the period
    between the stimulating RF pulses and the generation of the echo signal
  • The PEG dephases the spins along one axis in a known way. This changes the total
    signal recorded for each column of pixels.
  • The strength of the PEG changes with each repetition of the pulse sequence &
    produces a range of values for the summed signal in column of pixels.
A
  • By analysing how the summed signal
    changes with each different PEG, the signal
    from each pixel in the column can be
    calculated.
  • For a NxN matrix size, N different phase
    encode steps are required (N is typically
    128, 192 or 256)
34
Q

Frequency Encode Gradient (FEG)

Frequency Encode Gradient (FEG)
* The FEG is applied when the echo signal
is being detected (hence, it’s sometimes
called a “readout” gradient).
* The gradient alters the Larmor frequency
along one axis.
* The echo signal will now be detected at
different frequencies.

A
  • The frequency of the signal is related to
    the position of the voxel along the
    readout axis.
  • In this example, low frequency signal is
    received from voxels on the left of the
    patient, with high frequency signal from
    the right
35
Q

K-space

  • Each time a pulse sequence is operated a signal is sampled and stored in a computer.
  • The signal consists of a 128, 192 or 256 points of data (depends on the matrix size) at different frequencies.
  • Each sampled signal is stored as a single line of data in “k-space” (or “frequency-space”)
  • Each time the pulse sequence is repeated another line is filled in k-space.
A
  • K-space is the spatial frequency representation of the final MR image.
  • Low spatial frequencies are stored at the centre of k-space, high spatial frequencies are stored towards the boundary.
  • Performing the inverse fourier transform on k-space will convert it into a recognisable image.
36
Q

Centre of K space = low spatial frequency content = general contrast

Boundary of K space = high spatial frequencies - representation of edges fine detail

A
37
Q

Standard Pulse Sequences

  • A pulse sequence shows the order & timing of the gradients, RF pulses & data sampling windows that must be
    applied to produce an MR image.
  • The flip angles of the RF pulses and the timings of TR and TE largely affect signal strength & image contrast
  • Gradient field parameters will largely affect the spatial resolution
A
  • There are two standard types of pulse sequence
    – Spin echo (SE)
    – Gradient recalled echo (GRE)
  • All other MR pulse sequences are variations of SE or GRE
38
Q

Spin Echo Pulse Sequence

RF Pulse - Successive 90 and 180’s
Slice Selection
Frequency encoding gradient
With different Phase encoding gradient steps

You don’t measure initial FID after the 90 degree

You measure the ECHO after the 180 pulse

A
39
Q

Spin Echo Pulse Sequence

  • 90 - TE/2 - 180 - TE/2 - Echo
  • Initial 90o pulse creates a FID signal which will decay via T2*
  • The 180o pulse inverts the spin population
  • The spins “rephase” to create an echo at time TE after the 90o pulse
  • The rephasing 180o pulse compensates for magnetic field inhomogeneities
  • Can generate T1, T2 or PD weighted images
A

180-degree pulse in SE

40
Q

180-degree pulse in SE

A

T2* vs T2 Decay in Spin Echo

41
Q

Spin Echo Pulse Sequences

Pros
➢ Predictable image quality (SNR & CNR)
➢ Can generate T1, T2 or PD images
➢ Insensitive to Bo inhomogeneity

A

Cons
➢ Long acquisition times
➢ Acq Time = No of excitations x No of phase encoding steps x Repetition time (TR)
➢ (128x128 matrix needs 128 phase encoding steps)
➢ Use of 90o and 180o pulses results in high RF power deposition (high SAR)

42
Q

Variations of Spin Echo

A
43
Q

Turbo / Fast Spin Echo (FSE)

A
44
Q

Inversion Recovery (IR)

Inversion Recovery is a SE pulse sequence preceded by a 180o pulse
* 180 - TI - 90 - TE/2 - 180 - TE/2 - Echo

Inversion time DEPENDS ON T1

A
45
Q
  • After the inverting 180o pulse, Mz starts to regrow along the z-axis
  • After a time, TI = 0.693 x T1, Mz will pass through zero
  • A 90o pulse at this time will not generate an MR signal because there is no
    magnetisation vector to flip into the transverse plane.
A

FLAIR = Fluid attenuated inversion recovery

Water has a LONGER T1 so you FLAIR takes longer than for

Fat = STIR = Short Tau = Inversion time is SHORT for fat

46
Q

STIR & FLAIR
* Can chose TI to nullify the signal from any

tissue (occurs when TI=0.693 x T1)
* Short Tau Inversion Recovery
– “STIR”, TI ~160ms
– Nullifies the signal from fat

  • Fluid Attenuated Inversion Recovery
    – “FLAIR”, TI ~2200ms
    – Nullifies the signal from CSF
A
  • STIR advantages
    ➢ T1 or T2 weighted images
    ➢ Fat suppression
    ➢ Variable contrast possible (controlled by TI)
  • FLAIR advantages
    ➢ Suppression of CSF gives better visualisation of white matter
    ➢ Improved contrast in brain & spinal cord
  • Disadvantages
    ➢ Long acquisition times
    ➢ Increased SAR due to use of 180o RF pulses
47
Q

Gradient Recalled Echo (GRE)

  • Use a bipolar gradient to create an echo (usually along the frequency encode axis)
  • RF pulse – Bipolar gradient – Echo
  • GRE uses small flip angles (RF pulse < 90o) & short TRs
A
  • Fast imaging sequences
  • Susceptible to inhomogeneities in Bo
  • T1 and PD weighting is possible, but more difficult to
    predict than for SE
  • T2*-weighted images (T2 not possible)
  • FLASH, FISP, GRASS
48
Q

BIPOLAR FREQUENCY GRADIENT TO DEPHASE AND REPHASE

A
49
Q

Gradient recall echo
- Signal proportional to T2*

A
50
Q

Reduced Flip Angles

More along Mz to flip for short TRs

A
51
Q

GRE Image Weighting

A

Advantages
➢ Fast sequence with T1, T2* or PD weighting
➢ Sensitive to blood flow (angiography)
➢ Low power absorption (low SAR) due to small flip angles

Disadvantages
➢ SNR lower than SE
➢ T2 weighting not possible
➢ Susceptibility artifacts can lead to signal loss

52
Q

MR Angiography – Time of Flight

  • TOF angiography uses GRE with short TR and short TE
  • Signal from stationary protons in a slice is saturated, whilst spins flowing into the imaging slice are unsaturated and return a large signal
A
  • Blood flow perpendicular to the slice returns the highest signal
  • Slow blood flow or oblique blood flow across the slice will become saturated & produce a smaller signal
  • Angiograms are produced using a “maximum intensity
    projection” (MIP) algorithm from a stack of 2D slices
53
Q

Phase Contrast Angiography

  • A GRE sequence with a bipolar gradient is applied along the axis of a blood
    vessel to change the phase of flowing blood relative to stationary tissue
  • The phase of the signal from stationary spins is unchanged by a bipolar
    gradient, while the signal from flowing blood will have a phase shift
    dependent on the velocity and direction of blood flow.
A
54
Q

MR Angiography

A
55
Q

Chemical Shift Artifact

Water - less shielded - experiences a stronger field

A
56
Q

Chemical Shift Artifact

The difference in resonant frequency between protons in water and fat can cause misplacement of signal along the frequency encode axis. This is because the resonant frequency is directly related to position along this axis. This chemical shift artifact is more apparent at higher field strengths (Bo)

A
57
Q

Magnetic Susceptibility Artifacts

When an object is placed in a magnetic field, it’s atoms will interact with
it and this can change the strength of the field in the material.

The extent to which a material becomes magnetised is called its magnetic susceptibility.

A

There are 3 types:

➢ Diamagnetic – these materials have negative susceptibility & slightly reduce the field strength in the object (eg soft tissues & plastics)

➢ Paramagnetic – these materials produce an increase in field strength within the object due to interactions with unpaired electrons, (eg gadolinium and deoxyhaemoglobin)

➢ Ferromagnetic – these materials cause a very large increase in field strength in the object (eg, iron, steel, cobalt and nickel)

  • At the boundary between two tissues with different susceptibilities there
    will be LOCAL DISTORTIONS in Bo which will rapidly dephase the MR signal
    leading to SIGNAL VOIDS.
58
Q
  • At the boundary between two tissues with different susceptibilities there
    will be LOCAL DISTORTIONS in Bo which will rapidly dephase the MR signal
    leading to SIGNAL VOIDS.
A
59
Q

Magnetic Susceptibility

All tissues interact with the main magnetic field. This leads to local changes in
magnetic field strength within the tissue.

Local changes in magnetic field strength cause rapid dephasing of the MR signal,
leading to signal voids.

A
60
Q

Motion Artifact

Ghosting due to respiratory motion. These artifacts are typically observed along the phase encoding axis. Multiple PEGs are applied over a period of time, during which it is assumed the object remains unchanged.

A
61
Q

MR Hazards

Static field strength, Bo
* Temporary effects (metallic taste, dizziness)
* Projectile risk for ferromagnetic materials (those containing Fe, Ni & Co)
* Surgical clips, screws etc may move

A

Rapid gradient switching
* May cause peripheral nerve stimulation
* The control & programming of pacemakers may be adversely affected
* Acoustic noise levels

62
Q

MR Hazards - RF Fields

Specific Absorption Rate (SAR) = RF power absorbed per kg of tissue

  • SAR is proportional to (flip angle)2
A
63
Q

MR Safety Guidance

  • No legislation for MR safety
  • “Safety Guidelines for Magnetic Resonance Imaging Equipment in
    Clinical Use” – MHRA (2015) -
    https://www.gov.uk/government/uploads/system/uploads/attachment_data/fi
    le/476931/MRI_guidance_2015_-_4-02d1.pdf
  • Appoint an MR Responsible Person
A
  • Labelling of objects as MR safe, MR conditional etc
  • Restrict access to MR Environment where fringe field
    >0.5mT
  • Screen patients & visitors before entry to MR suite
64
Q

MR Hazards - Contrast Agents

  • Gadolinium salts are paramagnetic & reduce T1 relaxation times
  • Improve contrast on T1-weighted images
  • Gd-DTPA eg. Magnevist, Omniscan
A
  • Linked to Nephrogenic Systemic Fibrosis (NSF)
  • Not advised for GFR < 30ml/min/1.73m2
  • Pregnancy – Gd may accumulate in amniotic fluid. Must risk assess each individual case.
  • Breastfeeding - no need to stop
65
Q

Fringe field around an MR Scanner

Need to control access to areas where fringe field ≥ 0.5mT
Fringe field is not usually a problem for permanent magnets, but
will need to be actively shimmed for superconducting systems

A

*FRCR1 Physics (19 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2025 Bold Learning Solutions. Terms and Conditions