BASICS-AN OVERVIEW

 In human body hydrogen is abundant and it is randomly

oriented.

It consist of H+ion highly sensitive to the magnetic field.

S
When these randomly arranged protons of human body
brought to an external magnetic field some align towards the
magnetic field and other opposing the field.
Thus a net magnetization is produced.

 Net magnetization
in the longitudnal axis
Net magnetization thus produced!!!!!!

MRI

MAGNETIC FIELD
 PRECESSION

• In the presence of an external magnetic field, B0, the magnetic

moment will precess around the magnetic field like a spinning
top
B0 Precessional path

0 = Larmar frequency

Gyromagnetic ratio
Bo = strength of external magnetic field

0 =B0
 RESONANCE

 When two substances are in same frequency we can

say that the substances are in resonance.
 Change in energy of nucleus by the application RF
pulse is known as resonance.
 Thus the applied RF energy is given to the protons
which will flip the magnetization to the transverse
plane.
When the RF pulse is stopped the protons release energy and
comes to its original position .
This released energy is collected by the receiver coil and it is
processed and displayed as MR image.
 INTRODUCTION

• Parameters which are controlled and adjusted by the

operator .

• These changes influence image quality as well as

their sensitivity to pathology.
 Image contrast- factors

Primary factors Secondary factors

T1,T2,PD S/N ratio

TR, TE C/N ratio
Flip Angle Spatial Resolution
TI Coverage
FOV Acquisition Time
NEX
BANDWIDTH
SLICE THICKNES &INTERSLICE GAP
 T1 RELAXATION(SPIN-LATTICE)

• It is the time for 63% nucleus to return to the

lower energy state or longitudinal magnetization
after 90” RF pulse
 T1 WEIGHTED IMAGE

An image created using T1 relaxation or longitudinal relaxation

• using short TE and short TR times

• whose contrast and brightness are predominately determined by T1.

• The amount of T1 weighting controls by TR

TE-11
TR-468
 T1 Relaxation as a function of TR

Mz Fat W hite m atter

G ray m atter

63 %

C SF

t in ms Short T1 Long T1
240 680 809 2500 TR-468
TE-11
 T1 Constants (in ms)

TISSUE 1T 1.5T

Fat 240 320

Muscle 730 860
White matter 680 780
Gray matter 810 920

CSF 2500 3000

 T2 RELAXATION(SPIN – SPIN )

• The Time for a tissue to decay to 37% of its original

transverse magnetization.
• True T2 relaxation by 180” refocusing RF pulse .

T2

Time
 T2- WEIGHTED IMAGE

• An image created by T2 Relaxation or transverse relaxation

• whose contrast and brightness are predominantly

dependent by T2 relaxation .

• T2 occurs faster than T1 relaxation

TR-5580
TE-110
 T2 Contrast

Long T2 Tissue
Short T2 Tissue

Time
 Pixel containing protons with long T2 time.
with long T2 time. (eg Water)

M xy

C SF

G ray m atter
37 %

W hite m atter

Fat t in ms TR-5580
84 92 101 1400 TE-110
 T2* Decay

Decay of the FID following the RF excitation pulse.

Faster than T2 decay as it is combination of
1.T2 Decay itselF
2.Dephasing due to magnetic field inhomogenity

T2 Constants (in ms At 1.0 Tesla

Fat
85
Muscle
45
White matter
90
 PROTON DENSITY WEIGHTED IMAGE

• The concentration of mobile Hydrogen atoms within a

sample of tissue.
• Resulting in an image dependent primarily on the
density of protons in the imaging volume.
• In PD sequences we use short TE and long TR

TR-2700
TE-98
WEIGHTING TR TE

T1 SHORT( 450-850ms) SHORT(10-30ms)

T2 LONG (>2000ms) LONG(> 60ms)

PD LONG SHORT
 TR (Time of repetition)

It is the time in between the application of one RF pulse to

the application of next RF pulse.
Measured in milliseconds.
 TR determines T1 relaxation.
What happens if we or TR

Increasing TR………
It increases SNR.
Increase more coverage.
Increase proton density and T2 weighting

Decrease T1 weighting.
Increase scan time.
Decreasing TR..

Decreases SNR.
Increase TI weighting
Decrease scan time.
Decrease proton density and T2 weighting
Decrease coverage
 TE(time to echo)

It is the time difference between the application of the

excitation pulse to the peak of the signal.
Measured in milliseconds.
 What happens if or TE?

INCREASE IN TE…..
• Increase T2 weighting.
• Increase dephasing thus decrease SNR
• Decrease number of possible slice
• No change in scan time

DECREASE IN TE….

Increases SNR
Increases coverage
Decreases T2 weighting increase T1 or PD
No change in scan time
 What happens if or TE?

TR-3000 TR-3000 TR-3000

TE-22 TE-88 TE-242
 FLIP ANGLE

 It is the angle through which the net magnetization is flipped

towards the XY plane.

90° 180°
 FLIP ANGLE
The flip angle controls the transverse magnetization that is
created which induces a signal in coil

The lower the flip angle the lower the SNR.

More the flip angle more T1 weighted contrast image.

The maximum amplitude is created with flip angles of 90

degree.
 Inversion time
• It can suppress various tissues by selecting appropriate TI.
TI = 0.693xT1
• STIR(Short T1 Inversion Recovery) allows fat suppression by
selecting T1 value
.693x200ms = 140ms
• FLAIR(Fluid-attenuated Inversion Recovery) allows
T1-2500
fluid suppression by selecting T1value TR-8000
TE-108
.693x3600ms = 2500ms

Disadvantages
•Decrease SNR
•Decrease coverage
T1-160
TR-5310
TE-44
In.angle-180” STIR FLAIR
NEX -Number of Excitation or Acquisition
NEX means: number of times scan is repeated or
number of times data is collected with the same
amplitude of the phase encoding steps.

• NEX controls amount of data collected in each line

of k space.
• Doubles the NEX double the data stored in k space.
• Increasing the NEX reduces the motion artifact.
•Noise is randomly oriented in each time.

NEX by a factor 2 = noise by a factor 2

 SIGNAL TO NOISE RATIO

SNR is the ratio of the amplitude of the signal received by the coil to
the average amplitude of the noise.
 SNR is given by the equation

SNR ά (voxel volume) Ny * Nz * NEX

BW
Ny & Nz =phase encoding, frequency encoding steps.
NEX =number of excitation pulses.
BW =band width.

There for SNR is depends on:

1. Voxel volume = X*Y*
2. NEX
3. Number of phase encoding steps
4. Bandwidth
 Signal to Noise Ratio

low signal
high signal
low SNR
high SNR

frequency frequency
 VOXEL VOLUME


X=pixel in x direction,
Voxel volume = X*Y* Y=pixel in y direction
slice thickness

Voxel volume= Protons to spin= signal

 SLICE THICKNESS
According to this formula a change in slice thickness changes the
voxel size.

Doubling the slice thickness doubles the voxel size and so as

increase in SNR.
Thick slice have better SNR
They have reduced detail because of the effect of volume
averaging.
Often 5 mm slice provides excellent detail resolution
while maintaining excellent image quality.
 Number of phase encoding
steps(Ny)
PHASE ENCODING - the process of locating an MR signal by
altering the phase of spins in one dimension with a pulsed magnetic
field gradient along that dimension prior to the acquisition of the
signal.

It follows the same concept as like NEX

Ny is doubled increase SNR by 41% (2)

 SAMPLING
Conversion of the Analog signal to a series of digital values by
measurement at a set of particular times; this utilizes the analog to
digital converter.

If the rate of sampling is less than twice the highest frequency in

the signal,aliasing will occur.
1. Phase Oversampling
Oversampling is the increase in data to avoid aliasing
and wrap around artifacts.
Acquired k-space frequency data is not sampled density enough.
2. under sampling
Undersampling is the decrease in data to increase image acquisition
speed
 BW-Band Width

It is the range of frequency that are sampled during the

application of readout gradient.
• An inverse relation exists between bandwidth and SNR
i.e on decreasing BW=2 => SNR improves by a factor of = 2

Decreased BW Causes

•Increased SNR
•Increased Chemical Shift Artifacts
•Longer TE less signal due to more T2 decay(BW=1/Ts= Ts/Nx)
•Decreased No of Slices. (# of slices á TR/TE ratio)

# of slices=TR/(TE+Ts/2+To)
 Band Width –contd…

Bandwidth is the width, measured inHertz of a frequecy range in

which the signal's fourier transform is nonzero.
1. Receivers bandwidth
It is the range of frequencies accepted by the receiver
to sample the MR signal. SNR = 1/ (rBW).
2. Transmit bandwidth
The Transmit bandwidth refers to the RF excitation pulse required
for slice selection in a pulse sequence.
slice thickness proportional to the bandwidth of the RF gradient
strength
Lowering the pulse bandwidth can reduce the slice thickness
SNR can be increased by:
1. USE LONG TR AND SHORT TE.
2. A FLIP ANGLE OF 90.

3. A WELL TUNED AND CORRECTLY SIZED COIL.

4. A COARSE MATRIX.

5. A LARGE FOV.

6. THICK SLICES.

7. USE NARROW BANDWIDTH.

8. AS MANY NEX AS POSSIBLE

 FOV – field of view

 Defined as the size of the two or three dimensional spatial

encoding area of the image.

 Usually defined in units of mm².

 The FOV is the square image area that contains the object of
interest to be measured.

The smaller the FOV, the higher the resolution and the smaller
the voxel size but the lower the measured signal.

1.PHASE FOV=FOV in the phase encoding dorection

2.READ FOV =FOV in the frequency encoding direction
Fov= resolution = voxel size= signal
 INTERSLICE GAPS

Slice gap describes the amount of space between two

slices.
• An ideal slice profile should provide a
continous slices without gap.

• An unavoidable gap is created between two

slice

• So to avoid this gap we place slices as closer

which leads to cross talk.
 The overlapping area may show signals of both slices.
In order to get rid of cross talk we should use interslice gap
usually slice gap of 10 to 20 percentage is use in imaging.
Increase of slice gap results in less cross talk and increased coverage.
 CONCATENATION/INTERLEAVING

A measurement parameter, used e.g. for a short TR, to be

able to measure more slices, and
to prevent cross talk in the case of short slice distance.
CROSSTALK
An artifact introduced into images by interference
between adjacent slices of a scan. This artifact can be
eliminated by limiting the minimum spacing between
slices.

concatenation
Increase in interslice gap causes:

• Increase in coverage

• Decreased cross-talk artifact

• Increase in SNR (reducing cross talk using effective TR)

• Decresed detection of small lesions(with in gap)

 CONTRAST TO NOISE RATIO

• The difference in SNR between two adjacent areas.

• Controlled by the same factors that affect SNR.

A contrast to noise ratio is a summary of SNR and contrast.

It is the difference in SNR between two relevant tissue
types.
(A and B): CNR = SNRA – SNRB
 CNR = Contrast
Noise

CNR = SA-SB
Noise
 SPATIAL RESOLUTION

SPATIAL RESOLUTION - the ability to define minute adjacent

objects/points in an image, generally measured in line pairs per mm
(lp/mm).
Controlled by the voxel size
Can be increased by
 Thin slices
 Fine matrix
 Small FOV
TEMPORAL RESOLUTION
The shortest time duration between two events that can
be measured with an MRI experiment
Smaller fov causes…

Sag T1 weighted image

FOV=24cm

Sag T1 weighted image

Fov=12cm

Spatial resolution increase

SNR decreased
Thin slices gives….

Sag T1 weighted image

Slice thickness=10mm

Sag T1 weighted image

Slice thickness=3mm

Lower the slice thickness

Better the SNR,but increase
resolution
Fine matrix size…..

Matrix size=256*118

Matrix size=512*256
Improved spatial resolution
 COVERAGE
It is the distance covered by a multislice acquisition

It depends on :

Number of slices
slice thickness
Interslice gap
Number of slices=TR/(TE+Ts/2+To)
Then
Coverage =Number of slices *(slice thickness + gap)

Coverage= slice thickness, interslice gap, sampling time

 ACQUISITION TIME/SCAN TIME
The amount of time necessary to collect all of the data for a
particular sequence
The factors affect scan time are:

TR
Number of phase encodings
NEX
•Scan time(2D) = TR x Ny x NEX
•Scan time 3D = TR x Ny x Nz x NEX
•Scan time (FSE) = TR x Ny x NEX/ETL
To reduce scan time:
use shortest TR
coarsest matrix possible
reduce NEX maximum
 MAGNETIZATION TRANSFER
Discovered by Wolff and Balaban in 1989
MT relies primarily on free (bulk) water protons
Applying an off set RF pulse to the macromolecular protons, the
saturation of these protons is then transferred to the bulk water
protons. The result is net decrease in signal,but relative increase
In signal from pathological tissue.
The magnetization transfer ratio (MTR) is the difference in
signal intensity with and without MT.

WITHOUT WITH
MT MT
 TYPES OF COIL

• Receiver coils plays an extremely important role in

optimizing SNR
• Quadrature coil increase SNR

• Surface coil placed close to the area under exam increase SNR
 RF spoiling
spoiling is the method by which residual transverse
magnetization is destroyed (or cancelled) deliberately prior to
the application of the next RF pulse in the next TR, instead of
waiting for this magnetization to decay away.

Sequences may be RF spoiled or gradient spoiled, or both

 DISTORTION CORRECTION

 It is the inaccurate representation of the true size of an

object
 Or it is the displacement of displayed points within an image
relative to their true location in a slice.

DISTORTED IMAGE DISTORTION CORRECTED

 Image filtering
filtering is any process that alters the relative frequency
content.

This can be done with an analog (conventional

electrical) filter,

filtering can be carried out numerically on the digitized data.

Raw data can be filtered prior to the image calculation

 Triggering/gating

This method synchronize the heartbeat with the beginning of the TR

Where at the R wave is used as the triggerCardiac gating times
the acquisition of MR data to physiological motion in order to
minimize motion artifacts.

ECGgating techniques are useful whenever data acquisition is

too slow to occur during a short fraction of the cardiac cycle