MR physics and safety for fMRI€¦ · Wald, fMRI MR Physics Effect of Ear & Mouth Shim on EPI From P. Jezzard, Oxford B 0 Here is an unwarped EPI again without any passive shims,

Wald, fMRI MR Physics

HST.583: Functional Magnetic Resonance Imaging: Data Acquisition and AnalysisHarvard-MIT Division of Health Sciences and TechnologyCourse Instructor: Dr. Lawrence L. Wald.

MR physics and safetyfor fMRI

Lawrence L. Wald, Ph.D.

Massachusetts General HospitalAthinoula A. Martinos Center


Outline:Monday Oct 18 (LLW):

MR signal, Gradient and spin echoBasic image contrast

Wed. Oct 20 (JJ):Encoding the image

Monday Oct 25 (LLW):Fast imaging for fMRI, artifacts

Wed. Oct 25 (LLW):fMRI BOLD contrast

Mon. Nov. 1 (JJ):Review, plus other contrast mechanisms for fMRI (CBV, CBF)


Fast MR ImagingTechniques

• Why, introduction

• How: Review of k-space trajectoriesDifferent techniques (EPI, Spiral)

• Problems from B0 Susceptibility artifacts


Why fast imaging

Capture time course,(e.g. hemodynamic)eliminate artifact from motion (during encode.)


Magnetization vector durning MRRF

Voltage(Signal)

time

Mz

time

encode


Review of Image encoding,journey through kspace

Two questions:

1) What does blipping on a gradient do to thewater magnetization.

2) Why does measuring the signal amplitudeafter a blip tell you info about the spatialfrequency composition of the image (k-space).


Aside: Magnetic field gradient

Bo Gx x Bo + Gx x

Field from gradient coils

Uniform magnet Total field

zGx =∂Bz ∂xx


Step two: encode spatial info. in-plane

x

y

x

B F

ield

(w/ x

gra

dien

t) Bo

BTOT = Bo + Gz x

Freq.

Sign

alBo along z

υoυ

with gradient without gradient

“Frequency encoding”

v = γBTOT = γ(Bo + Gx x)


How does blipping on a grad. encodespatial info?

y

B F

ield

(w/ z

gra

dien

t) Bo

Bo τ

y1 y2

z

y

y1

y2

all y locsprocess at same freq.

all y locsprocess at same freq.

spins in foreheadprecessfaster...

Gy

υ(y) = γBTOT = γ Bo ∆y Gyθ (y) = υ(y) τ = γ Bo ∆y (Gy τ)



z

y

y1

y2

Bo

θ (y) = υ(y) τ = γ Bo ∆y (Gy τ)

yx yx

z

x y

z

υo

90°

yx

after RF After the blipped y gradient...

position y1 position 0 position y

z z



y

The magnetization vectorin the xy plane is wound intoa helix directed along y axis.

Phases are ‘locked in’ oncethe blip is over.


The bigger the gradient blip area, the tighter the helix

τ)θ (y) = υ(y) τ = γ Bo ∆y (Gy

y Gy

small blip medium blip large blip


What have you measured?Consider 2 samples:

1 cm

signal is as big as if no gradient

unifo

rm w

ater

no signal observed


Measurement intensity at a spatial frequency...

ky

1/10 mm

1/5mm

1/2.5mm

1/1.2mm = 1/Resolution

10 mm

kx


kx

ky1 / Resx

1 / FOVx

FOVx = matrix * Resx

Fourier transform


Sample 3 points in kspace

1/1.2mm = 1/Resolution

1/10 mm

1/5mm

1/2.5mm

More efficient!

t

kx

Gy

tGy

Frequency and phase encoding are the same principle!


Conventional “Spin-warp” encoding

“slice select”

“phase enc”

“freq. enc”(read-out)

RF

tS(t)

tGz tGy

Gx t

t

ky

a1

a2 kx

one excitation, one line of kspace...


Image encoding,

“Journey throughkspace”

The Movie…


kx

ky1 / Resx

1 / FOVx

FOVx = matrix * Resx

Fourier transform


Conventional “Spin-warp” encoding

“slice select”

“phase enc”

“freq. enc”(read-out)

RF

tS(t)

tGz tGy

Gx t

t

ky

a1

a2 kx

one excitation, one line of kspace...


“Echo-planar” encoding

RF

tS(t)(nograds)

tGz tGy

Gx

t

kx

ky

one excitation, many lines of kspace...

etc...

T2*

T2*


Bandwidth is asymmetric in EPI

ky

kx

• Adjacent points in kx have short ∆t = 5 us (high bandwidth)

• Adjacent points along ky are takenwith long ∆t (= 500us). (low bandwidth)

The phase error (and thus distortions) are in the phase encode direction.


Characterization of EPI performancelength of readout train for given resolutionor echo spacing (esp) or freq of readout…

RF tGz tGy

Gx

t

‘echo spacing’ (esp)

esp = 500 us for whole body grads, readout length = 32 msesp = 270us for head gradients, readout length = 17 ms


What is important in EPI performance?

Short image encoding time.

Parameters related to total encoding time:1) echo spacing.2) frequency of readout waveform.

Key specs for achieving short encode times:1) gradient slew rate.2) gradient strength.3) ability to ramp sample.

Good shimming (second order shims)


Susceptibility in MR

The good.

The bad.

The ugly.


Enemy #1 of EPI: local susceptibility gradients

Bo field maps in the head


What do we mean by “susceptibility”?

In physics, it refers to a material’s tendency to magnetize when placed in an external field.

In MR, it refers to the effects of magnetized material on the image through its local distortion of the static magnetic field Bo.


Ping-pong ball in water…

Susceptibility effects occur near magnetically dis-similar materials

Bo

Field disturbance around air surrounded by water (e.g. sinuses)

Field map (coronal image)

1.5T


Bo map in head: it’s the air tissue interface…

Sagittal Bo field maps at 3T


Susceptibility field (in Gauss) increases w/ Bo

1.5T 3T 7T

Ping-pong ball in H20:Field maps (∆TE = 5ms), black lines spaced by 0.024G (0.8ppm at 3T)


Other Sources of Susceptibility You Should Be Aware of…

Those fillings might be a problem…


Local susceptibility gradients: 2 effects

1) Local dephasing of the signal (signal loss) within a voxel, mainly from thru-plane gradients

2) Local geometric distortions, (voxellocation improperly reconstructed) mainly from local in-plane gradients.


1) Non-uniform Local Field Causes Local Dephasing

Sagittal Bo field map at 3T 5 water protons in different parts of the voxel…

z

slowest

fastestx

y

z90°

T = 0 T = TE


Local susceptibility gradients: thru-plane dephasing in grad echo EPI

Bad for thick slice above frontal sinus…

3T


Thru-plane dephasing gets worse at longer TE

3T, TE = 21, 30, 40, 50, 60ms


Problem #2 Susceptibility Causes Image Distortion in EPI

To encode the image, we control phase evolution as a function of position with applied gradients.

Local suscept. Gradient causes unwanted phase evolution.

The phase encode error builds up with time. ∆θ = γ Blocal ∆t

Field near sinus

y

y


Susceptibility Causes Image Distortion

Field near sinus

y

y

Conventional grad. echo, ∆θ α encode time α 1/BW


Susceptibility in EPI can give either a compression or expansion

Altering the direction kspace is transversed causes either local compression or expansion.

choose your poison…

3T whole body gradients


Susceptibility Causes Image Distortion

Echoplanar Image, ∆θ α encode time α 1/BW

3T head gradients

z

Field near sinus Encode time = 34, 26, 22, 17ms


EPI and Spirals

kx

ky

Gx

Gy

kx

ky

Gx

Gy


EPI SpiralsSusceptibility: distortion, blurring,

dephasing dephasing

Eddy currents: ghosts blurring

k = 0 is sampled: 1/2 through 1st

Corners of kspace: yes no

Gradient demands: very high pretty high

EPI and Spirals

EPI at 3T Spirals at 3T(courtesy Stanford group)


Nasal Sinus

B0

and the brain and head would be roughly positioned as seen. The B0 inhomogeneity in the inferior frontal cortex is similar to that observed experimentally.


Nasal Sinus + mouth shim

B0

In this preliminary study, we utilise a strongly diamagnetic material, in which case the shim can be placed inferior to the nasal sinus,in the roof of the mouth for example. The two B0 lobes, from shim and sinus, overlap to a much higher degree improving B0homogeneity in the inferior frontal cortex.

As the susceptibility of the diamagnetic shim is increased, so the B0 homogeneity in our region of interest improves. Subsequentactive shimming of the brain removes residual large scale B0 variations.If we consider a line profile across the region, we can predict that….


Effect of Ear & Mouth Shim on EPI

From P. Jezzard, Oxford

B0

Here is an unwarped EPI again without any passive shims, and now with the mouth shim present in addition to 3 cm3 of PGplaced behind each ear. You can see, for instance in slice 7, the good recovery of signal in the temporal cortices. However, thepresent design of ear shim provides little improvement in the more inferior regions of the temporal cortices.


With fast gradients, add parallel imaging

Acquisition: SMA

SH

SENSE

Reconstruction:

Folded datasets+

Coil sensitivity maps

Reduced k-space sampling

{

Folded images ineach receiver channel

FOVk π2

=∆


(iPAT) GRAPPA for EPI susceptibility

Fast gradients are the foundation, but EPI still suffers distortion

3T Trio, MRI Devices Inc. 8 channel arrayb=1000 DWI images

iPAT (GRAPPA) = 0, 2x, 3x


4 fold acceleration of single shot sub-millimeter SE-EPI: 23 channel array

23 Channel array at 1.5T

With and without 4x Accel.

Single shot EPI,256x256, 230mm FOVTE = 78ms

Encoding with RF…

Extending the phased array to more channels:23 channel “Bucky” array for 1.5T


9 Fold GRAPPA acceleration 3D FLASH

9 minute scan down to 1 minute…

3D Flash, 1mm x 1mm x 1.5mm, 256x256x128

23 Channel array at 1.5TCan speed up encoding by an order of magnitude!


Whys stop at 32? 96 Channels on its way…

Receiver array: Andreas Potthast

Helmut array:Graham Wiggins


Questions, comments to:

Larry Wald

Documents

MR physics and safety for fMRI€¦ · Wald, fMRI MR Physics Effect of Ear & Mouth Shim on EPI From P. Jezzard, Oxford B 0 Here is an unwarped EPI again without any passive shims,