33
2/23/12 1 Ari Borthakur, PhD Assistant Professor, Department of Radiology Associate Director, Center for Magnetic Resonance & Optical Imaging University of Pennsylvania School of Medicine BMB 601 MRI slide 2 A brief history Isidor Rabi wins Nobel prize in Physics (1944) for for his resonance method for recording the magnetic properties of atomic nuclei. Felix Bloch and Edward Purcell share Nobel Prize in Physics (1952) for discovery of magnetic resonance phenomenon. Richard Ernst wins Nobel Prize in Chemistry (1991) for 2D Fourier Transform NMR. His post-doc, Kurt Wuthrich awarded a Nobel Prize in Chemistry (2002) for 3D structure of macromolecules. Nobel Prize in Physiology & Medicine awarded to Sir Peter Mansfield and Paul Lauterbur (2003) for MRI. Who’s next? Seiji Ogawa for BOLD fMRI?

BMB 601 MRI - Center for Magnetic Resonance & Optical · PDF fileBMB 601 MRI slide 2 ... Felix Bloch and Edward Purcell share Nobel Prize in Physics (1952) for discovery of magnetic

Embed Size (px)

Citation preview

2/23/12

1

A r i B o r t h a k u r, P h D

A s s i s t a n t P r o f e s s o r , D e p a r t m e n t o f R a d i o l o g y A s s o c i a t e D i r e c t o r , C e n t e r f o r M a g n e t i c R e s o n a n c e & O p t i c a l I m a g i n g

U n i v e r s i t y o f P e n n s y l v a n i a S c h o o l o f M e d i c i n e

BMB 601 MRI

slide 2

A brief history…

  Isidor Rabi wins Nobel prize in Physics (1944) for for his resonance method for recording the magnetic properties of atomic nuclei.

  Felix Bloch and Edward Purcell share Nobel Prize in Physics (1952) for discovery of magnetic resonance phenomenon.

  Richard Ernst wins Nobel Prize in Chemistry (1991) for 2D Fourier Transform NMR.

  His post-doc, Kurt Wuthrich awarded a Nobel Prize in Chemistry (2002) for 3D structure of macromolecules.

  Nobel Prize in Physiology & Medicine awarded to Sir Peter Mansfield and Paul Lauterbur (2003) for MRI.

  Who’s next? Seiji Ogawa for BOLD fMRI?

2/23/12

2

slide 3

Outline Slide

# 3

  Hardware: !  MRI scanner !  RF coil !  Gradients

  Image contrast   Software:

!  Pulse sequences !  Fourier Transform

slide 4

How does it work?

http://www.med.harvard.edu/AANLIB/cases/case17/mra.mpg

2/23/12

3

slide 5

Original Concept*

1.  Rotating Gradient 2.  Filtered Back-projection

*Lauterbur, Nature (1973)

slide 6

Fourier Imaging*

1.  Three orthogonal gradients: !  Slice Selection !  Phase Encoding !  Frequency Encoding

2.  k-space !  Representation of encoded signal

*Kumar, et al. J. Magn. Reson. (1975)

2/23/12

4

slide 7

MRI scanners

19 Tesla! 4.7 Tesla! 1.5 Tesla!

To generate B0 field and create polarization (M0)

slide 8

MRI coils

Head coil

Mouse MRI platform

Torso coil To transmit RF field (B1) and receive signal

2/23/12

5

slide 9

Gradients

To spatially encode the MRI signal

slide 10

Image Contrast

Contrast in MRI based on differences in:   Relaxation times (T1, T2, T2

*, T1ρ) !  Local environment

  Magnetization = spin density !  Concentration of water, sodium, phosphorous etc.

  Macromolecular interactions !  Chemical-exchange, dipole-dipole, quadrupolar interactions

  Contrast agents !  Gadolium-based, iron-oxide, manganese

2/23/12

6

slide 11

Magnetization

Bitar et al. RadioGraphics (2006)

“spin” “spin ensemble” spin-up or

spin-down =

Boltzmann distribution

Net Magnetic Moment

slide 12

Energy levels

Larmor Equation

Energy gap

N+/N- = exp (-ΔE/kBT) Boltzmann distribution

2/23/12

7

slide 13

MR experiment

M0 aligned along B0 Mz=M0

RF pulse applied in x-y “transverse”

plane

M “nutates” down to x-y plane

Mz=M0cosα and

Mxy=M0sinα

RF off

Detect Mxy signal

in the same RF coil

slide 14

Equation of motion

Bloch Equation (simple form)

2/23/12

8

slide 15

Effect of RF pulse

The 2nd B field

Choosing a frame of reference rotating at ωrf, makes B1 appear static:

slide 16

Nutation

Lab frame of reference e.g. B1 oscillating in x-y plane

Rotating frame of reference e.g. B1 along y-axis

http://www-mrsrl.stanford.edu/ ~brian/mri-movies/

2/23/12

9

slide 17

After RF pulse-Relaxation

Spin-Lattice relaxation time=return to thermal equilibrium M0

Spin-spin relaxation with reversible dephasing

Spin-spin relaxation without “reversible” dephasing

T1>T2>T2*

slide 18

Signal detection

http://www-mrsrl.stanford.edu/ ~brian/mri-movies/

2/23/12

10

slide 19

How does MRI work?

Water ! Fat!4.7 ppm 1.2 ppm!

NMR signal!

?!=!

MR Image!

1) Spatial encoding gradients 2) Fourier transform

slide 20

Pulse Sequence (timing diagram)

Less MRI time/low image quality!

RF!

phase!

slice!

freq!!

90°!

TE!

Echo planar imaging!

Less

MRI

tim

e/lo

w im

age

qual

ity!

RF!

phase!slice!

freq!

90°!

TE!

Spin-echo!TR!

180°!

RF!

phase!slice!

freq!!

α (<90°)!

TE!

Gradient-echo!TR!

RF!

phase!slice!

freq!!

90°!

Fast spin-echo!TR!

180°! 180°! 180°!

2/23/12

11

slide 21

MRI

B0 M0 B1

Gz

Gx

Gy

slide 22

RF!

phase!

slice!

freq!

α!

TE!

Gradient-echo!

TR!

Slice Selection*

*Mansfield, et al. J. Magn. Reson. (1976) *Hoult, J. Magn. Reson. (1977)

2/23/12

12

slide 23

Mz

Slice Selection

B0

Mz

Mz

Mxy γGz•z RF

BWRF= γGz•Δz

Δz{

slide 24

Frequency Encoding*

RF!

phase!

slice!

freq!

α!

TE!

Gradient-echo!

TR!

*Kumar, et al. J. Magn. Reson. (1975)

a.k.a “readout”

2/23/12

13

slide 25

Frequency Encoding

Mz

γGx•x

BWread= γGx•FOVx

FOVx

slide 26

RF!

phase!

slice!

freq!

α!

TE!

Gradient-echo!

TR!

Phase Encoding*

*Kumar, et al. J. Magn. Reson. (1975) *Edelstein, et al. Phys. Med. Biol. (1980)

τpe

2/23/12

14

slide 27

Phase Encoding

γGy2•y

1/τpe= γΔGy•FOVy

FOVy

γGy1•y

γGy0•y

Each PE step imparts a different phase twist to the magnetization along y

slide 28

Traversing k-space

Spin-warp imaging*

RF!

phase!

slice!

freq!

α!

TE!

Gradient-echo!

TR!

ky

kx

*Edelstein, et al. Phys. Med. Biol. (1980)

2/23/12

15

slide 29

FT

k-space

freq phase

image

Fourier Transform

slide 30

Spiral Radial Echo-planar

phase!

freq!!

Echo planar! freq!

freq!!

Radial!

!freq!

freq!! Spiral!

Traversing k-space

2/23/12

16

slide 31

Fourier Transform

slide 32

Typical FT pairs

2/23/12

17

slide 33

K-space

The signal a coil receives is from the whole object:

S tx ,ty( ) = ρ x,y( )eiγ Gxtx ⋅x+Gyty ⋅y( )dxdy∫∫

kx, y = γGx, ytx, y

S kx ,ky( ) = ρ x,y( )eiγ kx ⋅x+ ky ⋅y( )dxdy∫∫

The image is a 2D FT of the k-space signal*:!

ρ x,y( ) = S kx ,ky( )e−iγ kx ⋅x+ ky ⋅y( )dkxdky∫∫

*Kumar, et al. J. Magn. Reson. (1975)

replace:!

K-space signal:!

slide 34

K-space weighting

High-frequency data=edges

Low-frequency data=contrast/signal

2/23/12

18

slide 35

What is wrong?

Original!

Freq. de-phaser!too large!!

*Pauly, http://www.stanford.edu/class/ee369b/tests/mt_sol.pdf

slide 36

What is wrong?

Original

ky step was too large or FOVy was too small!

2/23/12

19

slide 37

What is wrong?

Original!

Freq. gradient!amplitude was !too low!!

slide 38

MRI Examples

2/23/12

20

slide 39

3D Neuro MRI

slide 40

Susceptibility-Weighted Imaging (SWI)

2/23/12

21

slide 41

BOLD fMRI

slide 42

Phase Contrast MR Angiography

2/23/12

22

slide 43

TSE$Fat$Saturated$

TSE$TE$=$25$ms$

Resolution:200x200µm2

slide 44

7T Project: Hyper-polarized 3He MRI

K$Emami,$et$al.,$Mag$Reson$Med$2009$

2/23/12

23

slide 45

MRI

MRI “head coil”

Williams et al., PNAS 1994

7T Project: ASL-Perfusion MRI

Arterial Spin Labeling (ASL) technique

TAG

Imag

e

Con

trol

slide 46

ASL MRI & PET of Alzheimer’s Disease

ASL MRI Scan

18-FDG

PET Scan

CONTROL AD

Detre et al., Neuroimage 2009

2/23/12

24

slide 47

Sodium MRI of Inter-vertebral

Discs

slide 48

Motivation

  Lower Back Pain (LBP) is the most common ailment of the working-age adult.

  Affects > 4 million individuals each year in the US alone.

  Economic burden of $100 billion.

2/23/12

25

slide 49

Diagnosis of LBP-Discogram

1.  Inject saline to reproduce pain. 2.  Introduce dye to visualize defects.

slide 50

Mechanical function of spine-Pole Vault

•  Dynamic system"

•  Pole temporarily stores energy"

•  Releases energy delayed"

2/23/12

26

slide 51

Intervertebral Disc (IVD)

http://www.spineuniverse.com

slide 52

Hydrostatic System"

Mechanical function of IVD

2/23/12

27

slide 53

Disc Degenerative Disease

slide 54

Rationale for Sodium MRI

[PG] decreases with Degeneration

Urban et al Spine 1998

2/23/12

28

slide 55

Rationale for Sodium MRI

Urban and Maroudas Bioc. Biop. Acta 1979 Urban and Winlove, J. Magn. Reson. Imag. 2007

Sodium is a natural biomarker for PG

slide 56

How is [Na] related to [PG]?

  Fixed Charge Density (FCD) !  Side chains of PG !  [Na] correlated with [PG] through [FCD]

[Na] [FCD] [PG] detect early DDD

Maroudas et al., in Adult Articular Cartilage 1979 Lesperance et al., J. Orthop. Res. 1992

2/23/12

29

slide 57

Sodium MRI

What do you need? 1.  RF coil 2.  Transmit/receive switch 3.  Broadband capable scanner (all vendors) 4.  A short echo-time MRI pulse sequence

slide 58

Sodium MRI Ex Vivo

Can [PG] be quantified in the NP non-destructively?

1. Calibrate sodium MRI signal* from IVD.

y = 1.4x + 60.5

R2 = 0.99

0

100

200

300

400

500

0 50 100 150 200 250 300

calibration phantom [Na] (mM)

Sodiu

m S

ignal

Inte

nsi

ty (

a.u

.)

Shapiro et al., J. Magn. Reson. 2000

2/23/12

30

slide 59

Donnan Equilibrium

2. Convert [Na] in tissue to fixed charge density:

300mM

0mM

Lesperance et al., J. Orthop. Res. 1992

3. Convert FCD to [PG]:

slide 60

Sodium MRI Ex Vivo

Wang et al., Spine (2010)

•  Sodium maps in bovine discs

2/23/12

31

slide 61

Sodium MRI Ex Vivo

Wang et al., Spine (2010)

Sodium content vs. PG content

slide 62

Sodium MRI In Vivo

Wang et al., Spine (2010)

•  Sodium mapping vs. T2 MRI

2/23/12

32

slide 63

New 7T MRI!

  ultra-short echo pulse sequence

  TE/TR=200µs/25ms   2mm isotropic   15 minute acquisition

slide 64

Sodium MRI Movies

Axial Coronal Sagittal

2/23/12

33

slide 65

Sodium MRI Montage

  SNR ~26:1 in cartilage

slide 66

Take home… Slide # 66

  Magnet !  Create B0 !  Produce M0

  RF coil !  Transmit B1 field !  Detect Signal

  Image contrast !  Relaxation, concentration, interactions etc.

  Gradients !  Spatial encoding !  Signal in k-space

  Fourier Transform !  From k-space to image space

  Pulse sequences !  Traverse k-space !  Image Artifacts