3t mri bidmc department of radiology mr imaging and spectroscopy of the heart at 3t:technical...

3T MRI BIDMC DEPARTMENT OF RADIOLOGY

MR Imaging and Spectroscopy of the Heart at 3T:Technical Challenges


MR Imaging at 3T

2x S/N of 1.5 T

1/2 voxel size or

1/4 the acquistion time


MR Imaging at 3T

Technical Challenges:-Body RF Coil

Tissue Challenges:-T1’s get longer

Regulatory Challenges: SAR Bo^2


MR Imaging at 3T

Technical Challenges:-Body RF Coil

Why Have a body coil?

-critical for applications outside

the head

-homogeneous transmit coil for

Phased array studies

Original ResearchSensitivity and Power Deposition in a High-Field Imaging ExperimentDavid I. Hoult, MA, D, Phil *Institute for Biodiagnostics, National Research Council of Canada, Winnipeg, Manitoba, R3B 1Y6, Canada

Presented at the 7th Scientific Meeting of the ISMRM, Philadelphia, 1999

JMRI, 12:46-67,2000.

“SINCE THE EARLY DAYS of human imaging, it has been known that the electrical characteristics of tissue could adversely affect the fidelity of its image. Thus, Bottomley and Andrew ([1]) surmised that B1 field penetration effects could set an effective limit to the Larmor frequency of roughly 20 MHz, while independently but for the same reasons, Hoult and Lauterbur ([2]), in their paper on the signal-to-noise ratio (S/N) of the imaging experiment, suggested 10 MHz (0.25 T for protons) as a limit. Mansfield and Morris ([3]) adopted the same stance.”

Ultrahigh field (7T) magnetic resonance imaging and spectroscopy

Kâmil Uurbil, , a, Gregor Adrianya, Peter Andersena, Wei Chena, Michael Garwooda, Rolf Gruettera, Pierre-Gil Henrya, Seong-Gi Kima, Haiying Lieua, Ivan Tkaca, Tommy Vaughana, Pierre-Francoise Van De Moortelea, Essa Yacouba and Xiao-Hong Zhua

Magnetic Resonance Imaging21:1263-1281,2003

Better Spectra

7T vs. 4T: RF power, homogeneity, and signal-to-noise comparison in head imagesJ.T. Vaughan 1 *, M. Garwood 1, C.M. Collins 2, W. Liu 2, L. DelaBarre 1, G. Adriany 1, P. Andersen 1, H. Merkle 1, R. Goebel 3, M.B. Smith 2, K. Ugurbil 1

Magnetic Resonance in Medicine

Volume 46, Issue 1, Pages 24-30

4T 7T (7T/4T) 7T/4T(calc)


Poster #


3 Tesla Body Coil B1 Mapping• Resistive and Dielectric Properties of the Body Perturb RF

Uniformity at High Field

• Mapping of B1 in the body requires a fast, breathhold sequence

• Single shot FSE with different amplitudes of the excitation pulse was used

• Signal vs. amplitude was fit to approximately sinusoidal signal curve observed in phantoms

dielectric padsat 3T

pads nearcoil

pads nearpatient

dielectricshading

arrows indicate magnitudeand phase of B1

+ fieldcolor shows B1

+ field magnitude

GE Company ConfidentialTim Skloss and Armen Kocharian

Dielectric effects exist at all field strengths

These effects appear as non-uniformity in MR images

The effects are exacerbated at higher field strengths

The effects are exacerbated with multi-channel coils

Dielectric Effects – The Facts


So What Can Be Done to Minimize These Effects?

8-Channel Torso Coil

without any pad

8-Channel Torso Coil

with low conductivity pad


Low Conductivity Pad

20 millimolar solution of Manganese Chloride in distilled water.

(3.958 grams of Manganese Chloride (tetrahydrate) per liter of

solution)

A spiral volume coil for improved RF field homogeneity at high static magnetic field strength.

Alsop DC, Connick TJ, Mizsei G.

Magn Reson Med. 1998 Jul;40(1):49-54.

The Wave Equation Demands Spatial Variation of B Field

SpatialVariationOf RF

Short Wavelength Effect

Conductivity Effect

Birdcage vs. Spiral Coil

0°45°

90°135°

180° 0°45°

90°135°

180°

4 Tesla Spiral Head Coil Prototype

Designed for Whole Brain Imaging25 cm diameter, 30 cm length, Eight conductors

Distributed CapacitanceSeven 6.8 pf ceramic capacitors per conductor

Integrated RF ShieldMechanically connected, 32 cm diameter

Shield Current ReturnVaughan et al. MRM 32:206 (1994)

Coil Performance

High Q and Q ratioUnloaded Q 288, loaded Q 64

No tuning for load necessaryFrequency shift with load < 0.5 MHz

Excellent quadrature operationPolarity reversal dramatically reduced signal

Power deposition similar to birdcage100 mG B1 required 240W (CW)

Effect of Spiral on Uniformity

Spiral coil uniformity was clearly improvedCompares favorably with birdcage

Radial intensity variations consistent with theoryTheory assumes cylindrical symmetry

Low flip angle gradient echo imaging intensity=B2

Phase gradient less than expected66% of gradient expected for geometry

Radial Intensity Variation in 100% Isopropanol Phantom

Birdcage

Spiral

Human Head Imaging

Multi-slice low flip angle gradient echo imagingOxford Instruments 1 m 4 T magnetGE Horizon Echospeed HardwareTR/TE 500/3 , 10°, 32 kHz BW

Spiral coil reduces center brighteningIntensity more uniform than birdcage

Signal intensity drops off near top of headBoundary condition effect ?Independent of distance head is in coil

4 Tesla Head Imaging

Spiral Coil

Birdcage Coil

Summary

Spiral coil design improves RF homogeneity

No apparent penalty in power depositionFurther comparison studies required

Must compensate for dielectric boundariesVarying spiral pitch, radius with axial distanceExternal dielectric pads

Coil designs can overcome short RF wavelengths

Effect of External Dielectric


Increased FSE Slice Coverage

• Many multi-slice FSE protocols are limited by SAR even at 1.5 Tesla

• 3 Tesla multi-slice acquisitions take 4 times longer due to slice restrictions from the 4 x higher SAR

• Reduced flip angles can be used to make power identical to 1.5 T with only a small effect on sensitivity– D.C. Alsop, Magn Reson Med 37:176-184 (1997)


Sensitivity with Reduced Flip Angles

• Sensitivity drops only slowly with flip angle when tailored RF pulse trains are used for echo stability.

• Stimulated echo terms increase the effective T2 of the tissue but the images remain dominated by T2 contrast.

• Longer effective TE’s are required for the same T2 weighting.

For 90° pulses, SAR is reduced 4-fold but signal drops by just 14%


3 Tesla Reduced SAR FSE

• 90° asymptotic flip angles• 47, 3 mm slices in 3 acqs.• 16 ETL• 32 kHz BW• Flow compensation• TR 4000• 2 echoes• 256x256, 24 cm FOV• TE 12.4/112• 4 min 45 s total scan time


• Peripheral Gated Fastcard - SPGR• 19 Phases per 25 Second Breath Hold• 4 Element Cardiac Surface Coil Array

– GE R&D Center, Schenectady, NY

• Spatial Resolution: 1.3 x 1.5 x 8 mm

Cardiac Imaging Gradient Echo Imaging of the Heart


• End Diastole


• Mid Systole • End Systole

• Short Axis



• End Diastole • Mid Systole • End Systole

• Long Axis


Cardiac Imaging

2D FIESTA, Long and Short Axis

QuickTime™ and a decompressor

are needed to see this picture.

QuickTime™ and a decompressor

are needed to see this picture.

BLACK-BLOODFSE CARDIAC

IMAGING: 1.5T VS 3.0T

Robert L. GreenmanJohn E. ShiroskyRobert V. Mulkern

Neil M. Rofsky

• Published Studies– Gradient Echo - Signal ~ Sin– No Spin Echo (or FSE) Studies– Spin Echo - Signal ~Sin3

• B1 Heterogeneity– Conductive Effects (Signal Attenuation)– Dielectric Effects (Waveguide Effect) (or Resonant Cavity Effect)

FSE Black-Blood Imaging

• Changes in T2 Relaxation Times:– Tumors– Infarction– Cardiac Transplant Rejection

• STIR – Sensitive to Both T1 and T2 Changes– Suppresses Fat


• Blood Suppression– Minimizes Flow Artifacts

• Contrast– Vascular Walls– Endocardial Surfaces

• Double IR Pulse Sequence


FSE Black Blood Imaging

2 x RR

RectangularNon-Selective180 Pulse

Slice-SelectiveAdiabatic 180Pulse

Aquisition

RR

Double Inversion Black Blood Sequence

TI

1.5T Null Point = 625 ms


3.0T Signal at Calculated 1.5T IR

Time = -0.07 M0

Black Blood Imaging



3T Signal at Calculated 1.5T IR Time = -0.03 M0

• Double-IR FSE• Single Breathold• Matrix: 256 x 192• FOV: 40 cm• Slice Thick: 5 mm• Echo train Length (ETL): 24• Heart Rates: 45 - 75 BPM

Black Blood FSE Imaging 1.5T vs 3.0TMETHODS

• T2-Weighted

– Effective TE: 42ms (6th echo)

– TR variable 1.5 - 2.5 Sec

• STIR

– IR time variable for best fat suppression

– TR variable 1.5 - 2.5 Sec

• Cycled IR Pulses On and Off

• B1 Field Maps


• Body Coil Only• High-Pass Birdcage

• 1.5T Dimensions– 60 cm Diameter; 64 cm Long

• 3.0T Dimensions– 55 cm Diameter; 53 cm Long


B1 (RF) Field Maps

1.5 Tesla 3.0 Tesla

Black Blood FSE Imaging 1.5T vs 3.0T

Results


Results

T2-Weighted FSE Images

1.5T 3.0T


Results


Results - T2 W SNR

STIR FSE Images

1.5T 3.0T


Results


Results - STIR SNR

ROI MEASUREMENTS


BLOOD SUPPRESSION PERFORMANCE


Results


Results

BLOOD SUPPRESSION PERFORMANCE


Results

SNR (SIGNAL) UNIFORMITY


Results

SNR (SIGNAL) UNIFORMITY -P/A


Correlation of 23Na MR Imaging Findings with Cine, Late-Enhancement, and T2-weighted Findings

Note.—NA = not applicable.


RESULTS: All patients after subacute infarction and 12 of 15 patients with chronic infarction had an area of elevated 23Na signal intensity that significantly correlated with wall motion

abnormalities (subacute; r = 0.96, P < .001, and chronic; r = 0.9, P < .001); three patients had no wall motion abnormalities or elevated 23Na signal intensity. Only 10 patients in the subacute and nine in the chronic group revealed late enhancement; significant correlation with 23Na MR imaging occurred only in subacute group (r = 0.68, P < .05). Myocardial edema in subacute infarction correlated (r = 0.71, P < .05) with areas of elevated 23Na signal intensity but was extensively larger.


Sodium Imaging of the Heart

5 inch circular coil


Sodium Imaging of the Heart

8 inch circular coil

CHEMICALLY SELECTIVE PHOSPHORUS RARE

(FSE) IMAGING

CHEMICALLY SELECTIVE PHOSPHORUS RARE IMAGING

PHANTOM RESULTS

in vivo 31P RARE IMAGING PARAMETERS

• Modified FSE Sequence w/Chemical Selective Excitation

• Spatial Resolution: 4.7 X 4.7 X 25 mm (0.55 cm3)

• Scan Time: 4 Minutes/Metabolite Image (PCR or Pi)


Rest

Exercise

1H PCr Pi

FOREARM EXERCISE STUDY



FOREARM EXERCISE STUDY

1H/CONTOUROVERLAY

Pi/PCr Ratio

• Chronic High Glucose Levels Result in Functional Impairment of Circulation In Lower Extremities

• Ischemia and in 31P Metabolite Levels

• Neuropathy• Ulceration• Amputation

• Foot Muscle: Surrogate for Whole System


ISCHEMIA IN DIABETIC FOOT


1H PCr Pi



1H CONTOUROVERLAY

Pi/PCr Ratio



Alternative Non-Invasive Method for Assessment of Ischemia:MRS Chemical Shift Imaging

Scan Time: 34 MinutesResolution: 10 X 10 X 25 mm

(2.5 cm3)

Scan Time: 4 Minutes/ImageResolution: 0.47 X 0.47 X 25 mm

(0.55 cm3)

PCr Pi

3D RARE Pulse Sequence• Single Excitation

• Multiple Spin Echoes

• Readout Gradients Replace One CSI Phase Encode

31P Myocardial ImagingMethods - Pulse Sequence

Time

Readout

Phase Enc.

Slice Sel.

RF

ESP ESP ESP

ESP/ 2ESP ESP ESP

31P Myocardial Imaging in vivo Results

• 3D Acquisition - 2 adjacent slices

• 12.5 mm x 12.5 mm x 25 mm Voxels (4 cc)

• Scan time: 9 Minutes 40 Seconds

3t mri bidmc department of radiology mr imaging and spectroscopy of the heart at 3t:technical...

Documents

geometry slide

field magnitude slide

t 7t4t 7t4tcalc slide

t pads

t magnetic resonance

t technical challenges

liter of solution slide

acquistion time slide