Mri Pulse Seqs

MedicineAndHealthCancer 11,467 views 64 slides Feb 06, 2009
Slide 1
Slide 1 of 64
Slide 1
1
Slide 2
2
Slide 3
3
Slide 4
4
Slide 5
5
Slide 6
6
Slide 7
7
Slide 8
8
Slide 9
9
Slide 10
10
Slide 11
11
Slide 12
12
Slide 13
13
Slide 14
14
Slide 15
15
Slide 16
16
Slide 17
17
Slide 18
18
Slide 19
19
Slide 20
20
Slide 21
21
Slide 22
22
Slide 23
23
Slide 24
24
Slide 25
25
Slide 26
26
Slide 27
27
Slide 28
28
Slide 29
29
Slide 30
30
Slide 31
31
Slide 32
32
Slide 33
33
Slide 34
34
Slide 35
35
Slide 36
36
Slide 37
37
Slide 38
38
Slide 39
39
Slide 40
40
Slide 41
41
Slide 42
42
Slide 43
43
Slide 44
44
Slide 45
45
Slide 46
46
Slide 47
47
Slide 48
48
Slide 49
49
Slide 50
50
Slide 51
51
Slide 52
52
Slide 53
53
Slide 54
54
Slide 55
55
Slide 56
56
Slide 57
57
Slide 58
58
Slide 59
59
Slide 60
60
Slide 61
61
Slide 62
62
Slide 63
63
Slide 64
64

About This Presentation

No description available for this slideshow.


Slide Content

Single Echo Multi Echo
GRE SE GRE SESTE STE
Turbo*
FLASH
GRASS
SETurbo
STEAM
EPI FSEFIESTA
The PSD Family Tree

Sequences Across Vendors
Sequence Name GE Siemens Philips
Spin-echo MEMP,VEMP Spin-Echo Spin-Echo
Fast spin-echo FSE TSE TSE
Single-shot technique SSFSE HASTE Single Shot TSE
Coherent gradient-echo GRASS, GRE, FGR, FMPGR FISP, ROAST FFE
Incoherent gradient-echo (RF spoiled)SPGR, FSPGR T1 FFE
Incoherent gradient-echo (gradient spoiled)MPGR FLASH
Contrast-enhanced gradient-echo sequenceSSFP, DE FGR PSIF T2 FFE
Balanced coherent gradient-echo FIESTA, SSFP True FISP Balanced-FFE
Ultrafast gradient-echo FAST, GRASS, SPGR (IR/DE prep), IR FGRTurbo FLASH, 3D MP RAGETFE
Gradient and spin-echo GRACE GSE GRACE
Inversion recovery MPIR,TIR IR,TIR IR, IR-TSE, IR-TFE
Short T1 inversion recovery STIR STIR STIR
Phase-contrast sequence Phase Contrast Phase Contrast Phase Contrast
Parallel imaging technique ASSET IPAT SENSE

Options Across Vendors
Option Name GE Siemens Philips
Signal averaging NEX AC NSA
Partial averaging Fractional NEX Half Fourier Half Scan
Partial echo Fractional Echo Echo Partial
Rectangular field-of-view RFOV HFI RFOV
Off-center shifting slices Off Center FOV Shift, Offset Off Center Shift
Spacing between slices Spacing Distance Factor in % Slice Gap
Presaturation Spatial SAT SAT REST
Fat saturation FAT SAT, CHEM SAT FAT SAT SPIR, SPAIR, WaterSEL
Moving saturation pulse Walking SAT Travel SAT Travel REST
Gradient moment rephasing FC GMR FC
Respiratory compensation Respiratory Compensation, Respiratory TriggeringRespiratory Gated PEAR, Respiratory Trigger
ECG synchronization Cardiac Gated, Triggering ECG Triggered ECG Triggered
Delay after R wave Trigger Delay Trigger Delay Trigger Delay
Automatic bolus detection Smart Prep Care Bolus Bolus Track
Number of echoes ETL ETL, Turbo Factor TSE TF
Time between echoes Echo Spacing Echo Spacing Echo Spacing
Oversampling in frequency direction Always On Over-sampling Always On
Oversampling in phase direction No Phase Wrap Over-sampling Fold Over Suppression
Bandwidth Received Bandwidth Bandwidth Water/Fat Shift
Variable bandwidth VB Optimized Bandwidth Optimized Water/Fat Shift
Segmented k-space data acquisition Views per Segment Lines, Segments Views, Segments
Multislice imaging Multi Slice Multi Slice Multiple Slice
3D Imaging 3D 3D Volume 3D
Orientation scan Localizer Localizer, Scout Plan Scan, Survey

SI
slice selection gradients
FID
a RF "sinc" pulse - flip angle
4 msec
sampling time
dephase
Gradient-recalled-echo (GRE)
slice

SI
FID
90 RF "sinc" pulse
4 msec
sampling echodephase
Spin-echo (SE)
180 RF "sinc" pulse
sliceslice

Single Echo
GRE SESTE
Turbo*
FLASH
GRASS
SE Turbo
STEAM
Subsecond
Easy to run
Flexible
Popular
Poor SNR
Popular...
Conventional
Keyholed...
Seconds
More difficult
Very flexible
Popular

Multi Echo
GRE SE STE
EPI
Spiral
FSE FIESTA
Subsecond 10Hz
Difficult/expensive
Very flexible
Good SNR
Intense interest
Seconds to minutes
Easy, SAR
Very flexible
Good SNR
Very popular
Fast but sensitive
To field
inhomogeneities

K-space covers Frequency and Phase
Frequency
Phase
Sequences:
GRE
SE
FSE
Turbo GRE
EPI
Spiral

Sampling
VB
1/2NEX
1/2 Echo
Key Hole

k-Space and MRI Physics
Phase
Frequenc
y
hi
lo
+
-
hi
FT
lo
"sample spin echo"
slice
phase
frequency
90 rf 180 rf
Frequenc
y
Phase

The MR Sequence Determines How k-space
is Sampled...
90 180
view 1
192
The "spin-echo"
128
64
256
view 1
view
256
Frequency
Slice selection

High-Speed MRI Families
GRE SE
Turbo
single
echo
multi
echo
EPI
multi
echo
FSE
~ 1 sec
ez to do
~ 32 msec
strong gradients
needed...
10 sec - 6 min.
SE tissue contrasts

High-Speed GRE Family
GRE
SPGR
spoiled
Refocused GRE
GRASS MPGR
Appears T1-wt
Useful for GM-WM
3D GRE uses SPGR for
anatomy - MP-RAGE
Appears T2-wt.
Useful for MRA, flow, CSF, MS

GRE vs. EPI
Slice
Frequency
Phase
Partial flip
Partial flip
echo
echo
echo echo
echo

GRE vs. EPI
One view every TR...
All views in one TR!
Note where the "center
echo" - this is the key to
TEf - the effectve TE!

Alternate K-Space samplings
MR projections MR “spirals”

Spiral Scanning
Segmented sampling
Done with less than 1 gauss/cm
Non-linear ADC sampling
Incredible applications (fast, flow sensitvie)
slice
x-gradient
y-gradient

LAD
Spiral-scan
Meyer et al.
Stanford

Transverse Magnetization builds up for
short TR values in Gradient-echo MRI!!
REFOCUS OR SPOIL
REWIND OR REFOCUS - GRASS, MPGR
SPOIL - SPGR

SE
FOV 16, 4 NEX, 128X256, 5 mm
MPGR 50/10

SI
slice selection gradients
FID
a RF "sinc" pulse
4 msec
sampling time
dephase
TR 10-100
gradient-echo still dephasing for gradient-echo still dephasing for
4XT2*4XT2*
Transverse magnetization builds up Transverse magnetization builds up
from view to view.from view to view.

SI
slice selection gradients
FID
a RF "sinc" pulse
4 msec
sampling time
dephase
"rewinds" phase
from view to view
REWIND OR REFOCUS

SI
slice selection gradients
FID
a RF "sinc" pulse - VARY PHASE OF EACH PULSE
4 msec
sampling time
dephase
"rewinds" phase
from view to view
SPOILED GRASS
KILLER

Contrast in GRE
TR 200
TE 15
MPGR
5
o
45
o
75
o
For long TR:
Increase in
flip adds
T1-wting.....

Contrast in GRE
TR 200, Flip 30, MPGR
For long TE
in GRE:
Increase in
TE adds
T2-wting.....
and MS artifacts
6 ms
30 ms

Basic Rules of GRE Tissue Contrast
GRASS, MPGR - Gives T1, T2, T2* mixed contrasts
SPGR - Gives (nearly always) T1-weighted contrast

Basic Rules of GRE Tissue Contrast
SPGR - Gives (nearly always) T1-weighted
contrast
For best T1-weighting:
TR short 20- 100
TE short as possible < 10
Flips from 30-45

In General...
Small flip angles enhances proton density (PD).
Increasing flip leads to more T1-weighting.
Increasing TE leads to more T2, T2*-weighting.
Increasing TR, decreasing flip leads to more PD.
Mechanisms in Action
Steady-state transverse phase loss - T2, T2*
Longitudinal recovery - T1
FRE, magnetic susceptibility - FRE, MS

GRE - Mutants
"Fast" GRE - Done with short TR = "Turbo"
"Center out" - Do the center views first to minimize
T1 saturation...
"DE" - Driven equilibrium - does a 90 - 180 - GRE
to create a "T2- like" tissue contrast...
"IR" - Inversion recovery to give T1 weighting...
F/W phase - adjusts TE to make fat/water in phase
useful to help minimize lipids

SE vs. FSE
Slice
Frequency
Phase
90 180
echo
90180 180 180 180 180

sample
frequency
increment
phase
change
FSE k-space sampling is simply view by view

FSE Advantages
Short scan times!!
TR 2000, 512X512, 2 NEX = 34 minutes
Scan time = TR x #views x NEX
TR 2000, 512X512, 2 NEX , ET 16 = 2 minutes!!

The "effective TE" in FSE is determined by
where the "center views" (LEAST AMOUNT
OF PHASE ENCODING!) are collected...
Effective TE in FSE
FSE
90180echo
Phase
encoding
Where is the center view? What is the eff. TE??

Effect of TE
TR 5000
TE 85
TE 119 TE 136
TE 102
FSE
allows
long
TR's...
helps TE
effect by
reducing
T1
contributio
ns!!

FSE Issues
Total TR
8 echoes give
6 slices...
16 echoes give
3 slices...

SI
Time (msec)
17
34
51
68
85
102
136
153 msec
Long ET trains increase T2-weighting...
For longer ET trains,
the
later echoes contribute
greatly...
This is ET 16!!

Effect of TE
TR 5000
TE 85
TE 119 TE 136
TE 102
FSE
allows
long
TR's...
helps TE
effect by
reducing
T1
contributi
ons!!

Effect of ET
TR 4000
TE 102
ET 4
ET 16
ET 8
Increasing ET
increases
T2 effects...
By increasing the
contribution
from the later
echoes...

FSE has unique applicatons in the body..
TR 2000
TE 85
ET 16
128X256
1 NEX
BREATH
HELD!

FSE has
bright fat..
Virtues of
FATSAT...
TR 6000
TE 119

Chem Sat (FatSat)
1. Wide range of clinical advantages:
Anatomy free of lipids (or water!)
Better SNR (dynamic range is increased, as
is amplifier gain)
Definition of hyperintensity as fat/fluid.
Reduction of fat-enhanced respiratory
artifacts!!
NO chemical shift misregistration
VB now possible
2. Clinical roles:

Better depiction of joint fluids
Improved Gd enhancement

How does FATSAT work?
The chemical shift is due to the different proton
environments of water and lipids...
frequency
Water
Lipids
220 Hz
+
+
This chemcial shift is responsible for the
"Chemical shift artifact"

How does FATSAT work?
Apply a long (16 msec) rf pulse exactly at the
lipid resonance but miss the water resonance...
frequency
Water
Lipids
+
+
This chemcial shift is responsible for the
"Chemical shift artifact"
Bandwidth of rf pulse = 100Hz

How does FATSAT work?
Apply a long (16 msec) rf pulse exactly at the
lipid resonance but miss the water resonance...
The "sinc" pulse profile has a square bandwidth..
Time
Frequency
RF pulse length
related to RF bandwidth
center
frequency
bandwidth

How does FATSAT work?
Now that lipids are excited...
Crush phase coherence with a strong GRADIENT!
SAT RF "Crush"90 RF
180 RF
gradient
Excites only lipids
220 Hz from water
Spoils phase
coherence
Begin next view
with 90 - 180 - echo...

FATSAT SPGR, FOV 16, 4 NEX, 128X256, 5 mm
SAT off SAT on

Steady-state Free Precession - 1
Improved gradient capabilities -> ultrashort TRs
SSFP for
cardiac imaging,
flow imaging
T1/T2 quantification,
whole body imaging
SSFP is unique - both spin-echo and gradient-echo
In SSFP
A gradient-echo is acquired like FLASH or FISP
But, gradients are symmetrically balanced in slice,
phase-encoding, & read directions
No RF spoiling is implemented.

Steady-state Free Precession - 2
Coherences are maintained in successive TRs.
Transverse magnetization from one TR contributes to the
next TRs.
The RF echoes generated by the train of α pulses in the
steady state are then added to the gradient-echo
(assuming a uniform main field, this is the catch).
Signal is mostly from long-T2* and long-T2 tissues.
No spoiling, so no saturation effects as in spoiled
techniques, and high flip angles can be used in SSFP.
Balanced gradients also mean motion insensitivity.

Steady-state Free Precession - 3
As long as TR is shorter than T2 without RF spoiling, we
have a coherent steady state.
That is a combination of the longitudinal and transverse
components. It takes time to reach SS.
In steady state, the SSFP signal is T2/T1 weighted.
SSFP signal is a complicated function of parameters such
as TR, T1, T2, flip angle, and off-resonance angle β (β =
×
γ Δ
B0 ×TR).
Advantages: unique contrast, high SNR compared to
spoiled GRE & high imaging efficiency, but
sensitivity to off-resonance is a major limitation.

Figure 5-32 SSFP pulse sequence diagram. The gradients are balanced along all three axes
so that steady-state effects related to long-T2* species are emphasized. The gradient-echo
and the RF echo are superimposed at TE, and the gradient structure is motion insensitive.
Downloaded from: Clinical Magnetic Resonance Imaging, 3rd edition (on 3 April 2007 05:38 PM)
© 2007 Elsevier

Off-resonance Artifacts in SSFPOff-resonance Artifacts in SSFP
Off-resonance artifacts are usually bands in SSFP.
Main sources of off-resonance artifact
B0 inhomogeneity.
Static B0 field varies within an object.
Artifacts can be minimized by
careful shimming, high BW, shortest TR possible.
Approach to Steady StateApproach to Steady State
3 x T1 to reach steady state.
Long T1 tissues may show artifact.

FIESTA Example
T2 FSE FIESTA
T1 FSE

Reduced Scan time Data Acquisition Strategies
Fractional Echo
Partial Views
SMASH
Partial FOV
SENSE
Increased SNR Data Acquisition Strategies

SMASH - Simultaneous Acquisition of Spatial Harmonics
Sodickson & Manning, MRM 38(4):591-603 (Oct. 1997)
“Linear combinations of
simultaneously acquired signals
from multiple surface coils with
different spatial sensitivities to
generate multiple data sets with
distinct offsets in k-space.”
Huh?
Add signals from multiple coils
using information about the coil’s
spatial location

SMASH - Simultaneous Acquisition of Spatial Harmonics
Sodickson & Manning, MRM 38(4):591-603 (Oct. 1997)
S(k
x
,k
y
)= C(x,y)M(x,y) e
∫∫
-I(kx· x+ky · y)
dxdy
Usually C(x,y) = 1
For SMASH,
Construct coils with
sinusoidal sensitivities,
Like a gradient shift.
5 lines per readout with
4 spatial harmonics per readout

SMASH achieves a reduction in scantime, R, given by the number
of simultaneously acquired spatial frequency harmonics.
SMASH Procedure:
1. Determine sensitivity profile for each coil.
2. Determine the number of spatial harmonics that can be
generated using the coil array.
3. Acquire data from coil array - these are aliased component coil
images.
4. Determine weights for linear combinations of component coil
signals.
5. Form composite k-space signals corresponding to the spatial
harmonics.
6. Interleave the composite signals then Fourier transform.
Weakness: Measurement & manipulation of sensitivity profiles.

SENSE: Sensitivity encoding
Pruessmann, Weiger,Scheidegger,Boesigner. MRM 42(5):952-969 (Nov. 1999)
“Knowledge of coil
sensitivity implies information
about the detected MR signal
which may be used in image
generation.”
Coil 2Coil 1
Coil 4Coil 3
SMASH requires the
combination of coil sensitivity.
SENSE is the generalization
of SMASH for any geometry.

Aliasing: MRI data is collected in the frequency MRI data is collected in the frequency
domain, so objects outside of the FOV fold back domain, so objects outside of the FOV fold back
into the image.into the image.
Analog
Signal
Over-
sampled
Under-
sampled
f1
f2
f1

Aliasing or Wrap-around Aliasing or Wrap-around in standard coil seen
when
FOV is smaller than the object being imaged.
FOV xFOV x
FOV yFOV y
TrueTrue
positionspositions
AliasedAliased
positionspositions

SENSE: Sensitivity encoding
Pruessmann, Weiger,Scheidegger,Boesigner. MRM 42(5):952-969 (Nov. 1999)
Coil 2Coil 1
Coil 4Coil 3
Reconstruction of an image from N receiver coils:
Undersampled k-space from
each receiver (aliasing).
Undo signal superposition
caused by fold-over
(aliasing).
Undo signal superposition by
using weighting caused by
varied coil sensitivities.

FOV=24cm, 384x256, 5mm slice, TR/TE=4400/97.4Ef ms
EC=1/1, BW= 31.2 kHz, 2 NEX, VBW/TRF/Z512
WVU Twin-speed ACR Uniformity Slice: ASSET compatible FRFSE-XL/90
After ASSET cal. - ASSET turned on
without SCIC with SCIC
8 Channel Brain Coil

FOV=24cm, 384x256, 5mm slice, TR/TE=4400/97.4Ef ms
EC=1/1, BW= 31.2 kHz, 2 NEX, VBW/TRF/Z512
WVU Twin-speed ACR Uniformity Slice: ASSET compatible FRFSE-XL/90
After ASSET cal. - ASSET turned off
without SCIC with SCIC
8 Channel Brain Coil

Problem is that while the SNR as measured by Center Signal / stdev
of background is high, uniformity is poor. In a ROI of 500 mm
2
stdev
= 50, or 25 w/SCIC.
WVU Twin-speed ACR Uniformity Slice: ASSET compatible FRFSE-XL/90
with SCIC
8 Channel Brain Coil Quad Head Coil
SNR w/SCIC = 195, wout/SCIC = 128 SNR= 95

WVU Twin-speed ACR Uniformity Slice: ASSET compatible FRFSE-XL/90
Coil SignalSt.Dev.
Head 62211
8ch 67750
8chSCIC 62325
Measurements made in same
location - as shown in image.