Overview. Optimizing MRI Protocols (SNR, CNR, Spatial Accuracy) Image Contrast. Image Quality. Intrinsic (Tissue) Parameters



Similar documents
Overview. Optimizing MRI Protocols. Image Contrast. Morphology & Physiology. User Selectable Parameters. Tissue Parameters

MRI SEQUENCES. 1 Gradient Echo Sequence

GE Medical Systems Training in Partnership. Module 8: IQ: Acquisition Time

GE 3.0T NPW,TRF,FAST,F R NPW,TRF,FAST,F R

Musculoskeletal MRI Technical Considerations

GE Medical Systems Training in Partnership. Module 12: Spin Echo

5 Factors Affecting the Signal-to-Noise Ratio

Basic Principles of Magnetic Resonance

SITE IMAGING MANUAL ACRIN 6698

Table 11: Pros and Cons of 1.5 T MRI vs. 3.0 T MRI; Safety and Technical Issues, and Clinical Applications

Master s Program in Medical Physics. Physics of Imaging Systems Basic Principles of Magnetic Resonance Imaging V. Prof. Dr.

MGH Adult Diffusion Data Scanning Protocols

7/16/2010. Pulse Sequences and Acquisition Techniques for Breast MRI. Objectives. ACR Breast MRI Accreditation Program Launched May 2010

Diffusione e perfusione in risonanza magnetica. E. Pagani, M. Filippi

Kap 8 Image quality, signal, contrast and noise

Etude POPART'MUS MRI Component

3/30/2013. Disclosure. Advanced Neuro MRI: Imaging Techniques and Protocol Optimization. MRI, 35 year ago. MRI Today. Outlines

Chapter 2 NMR in Inhomogeneous Fields

MRI Physics for Radiologists

7/22/2011. Breast MRI: Pulse Sequences, Acquisition Protocols, and Analysis. Objectives. Challenges in DCE Breast Imaging

Medical Imaging. MRI Instrumentation, Data Acquisition, Image Reconstruction. Assistant Professor Department of Radiology, NYU School of Medicine

Generation and Detection of NMR Signals

MRI for Paediatric Surgeons

THEORY, SIMULATION, AND COMPENSATION OF PHYSIOLOGICAL MOTION ARTIFACTS IN FUNCTIONAL MRI. Douglas C. Noll* and Walter Schneider

NMR Techniques Applied to Mineral Oil, Water, and Ethanol

Magnetic Resonance Imaging

Cirrus 0.2T. MRI for Everyone. North America, Asia, Europe. contact:

REVIEW. Magnetic Resonance: An Introduction to Ultrashort TE (UTE) Imaging

High Field MRI: Technology, Applications, Safety, and Limitations. Brief overview of high-field MRI. The promise of high-field MRI

Proton magnetic resonance spectroscopy in the brain: Report of AAPM MR Task Group #9

32-Channel Head Coil Imaging at 3T

Clinical applications of MRI in radiation therapy. Jatta Berberat, PhD Kantonsspital Aarau

FDA Guidelines for Magnetic Resonance Equipment Safety

Glossary of MRI Terms

Electronic Supplementary Information

High Spatial Resolution EPI Using an Odd Number of Interleaves

In vivo stem cell tracking in the myocardium of small laboratory animals: Feasibility and acquisition strategies on clinical magnetic resonance

Accurate Multislice Gradient Echo T 1 Measurement in the Presence of Non-ideal RF Pulse Shape and RF Field Nonuniformity

Toshiba Excelart Vantage 1.5T MRI Tech Specs (Technical Specifications)

Pulsed Fourier Transform NMR The rotating frame of reference. The NMR Experiment. The Rotating Frame of Reference.

NMR for Physical and Biological Scientists Thomas C. Pochapsky and Susan Sondej Pochapsky Table of Contents

Technique and Safety of. by Pierluigi Castellone, Electronics Engineer Brain Products General Manager

How To Understand The Measurement Process

runl I IUI%I/\L Magnetic Resonance Imaging

E. K. A. ADVANCED PHYSICS LABORATORY PHYSICS 3081, 4051 NUCLEAR MAGNETIC RESONANCE

Nuclear Magnetic Resonance

Investigation of Magnetic Resonance Imaging Effects when Using Bone Conduction Implants

Advanced MRI methods in diagnostics of spinal cord pathology

Nuclear Magnetic Resonance and the Measurement of Relaxation Times of Acetone with Gadolinium

Lecture #7 (2D NMR) Utility of Resonance Assignments

Nuclear Magnetic Resonance

Reproducibility of T1 and T2 Relaxation Times Calculated from Routine MR Imaging Sequences: Phantom Study

The Magnetic Resonance Imaging Examination

Nuclear Magnetic Resonance (NMR) Spectroscopy

Introduction to Magnetic Resonance Imaging Techniques

ORIGINAL PAPER WATER FAT SEPARATION USING THE SINGLE ACQUISITION SANDWICH TYPE 3-POINT DIXON METHOD TO OPTIMIZE KNEE JOINT SCANS

BIOMEDICAL ULTRASOUND

mri : Physics For anyone who does not have a degree in physics Evert J Blink Application Specialist MRI

Characterization of small renal lesions: Problem solving with MRI Gary Israel, MD

NEURO MRI PROTOCOLS TABLE OF CONTENTS

510(k) Summary. This summary of 510(k) safety and effectiveness is being submitted in accordance with the requirements of 21 CFR 807.

Case Report: Whole-body Oncologic Imaging with syngo TimCT

Lecture 14. Point Spread Function (PSF)

Principles of functional Magnetic Resonance Imaging

Tetramethylsilane (TMS) Trimethylsilyl d 4. -propionic acid (TMSP) Dioxane. O - Na + Dimethylfura n. Potassium Hydrogen Phthalate. Sodium Maleate CH 3

THE WIDEST WIDE-BORE IN THE INDUSTRY

RF Coils... They ve Come a Long, Long Way

MRI. Chapter M. Contents M.1

Breast MRI Quality Control

Nuclear Magnetic Resonance (NMR) Spectroscopy cont... Recommended Reading:

Update: MRI in Multiple sclerosis

Rapid Combined T 1 and T 2 Mapping Using Gradient Recalled Acquisition in the Steady State

Physiological Basis of the BOLD Signal. Kerstin Preuschoff Social and Neural systems Lab University of Zurich

Magnetic Resonance Multiphase Flow Meter: Gas Flow Measurement Principle and Wide Range Testing Results

Pulsed Nuclear Magnetic Resonance An Experiment for UCSB s Advanced Laboratory

MEDIMAGE A Multimedia Database Management System for Alzheimer s Disease Patients

Artifacts caused by transcranial magnetic stimulation coils and EEG electrodes in T 2 *-weighted echo-planar imaging

The Hydrogen Atom Is a Magnet.

Unwarping Echo Planar Images Using CMTK 1 Release 1.2

4. It is possible to excite, or flip the nuclear magnetic vector from the α-state to the β-state by bridging the energy gap between the two. This is a

NMR Pulse Spectrometer PS 15. experimental manual

Super-Resolution Reconstruction in MRI: Better Images Faster?

Standardized MRI Protocol for Brain Tumor Clinical Trials. Benjamin M. Ellingson, Ph.D. Assistant Professor of Radiology at UCLA

Nuclear Magnetic Resonance Spectroscopy

NeuroImage. MP2RAGE, a self bias-field corrected sequence for improved segmentation and T 1 -mapping at high field

Phantom Test Guidance

The MRI Study Guide for Technologists

Nuclear Magnetic Resonance Spectroscopy

GUIDE TO SETTING UP AN MRI RESEARCH PROJECT

MRI QA Technologist s Tests

Hunting Bats. Diagnostic Ultrasound. Ultrasound Real-time modality

R/F. Efforts to Reduce Exposure Dose in Chest Tomosynthesis Targeting Lung Cancer Screening. 3. Utility of Chest Tomosynthesis. 1.

Transcription:

Overview Optimizing MRI Protocols (SNR, CNR, Spatial Accuracy) Geoffrey D. Clarke, Ph.D. University of TX Health Science Center San Antonio, TX Parameters Affecting Image Quality Manipulating Image Contrast (Standard) Classifications of Pulse Sequences Relationships Between Resolution & SNR MR Image Speed Manipulating Image Contrast (Advanced) Image Quality Image Contrast Image Resolution Signal-to-Noise Ratio Imaging Speed Image Contrast Basic image contrast is effected by the amplitude and timing of the RF pulses used to excite the spin system. More advanced methods may use gradient pulses (to modulate motion) and alter tissue properties with exogenous contrast agents Intrinsic (Tissue) Parameters Tissue Water Density (Proton Density) Longitudinal Relaxation Time (T1) Magnetization transfer rate coefficients Transverse Relaxation Time (T2) Chemical Shift (water vs. fat) Tissue Motion Macroscopic (flowing blood) Microscopic (apparent diffusion coefficient) Tissue Susceptibility (air, fat, bone, blood) Extrinsic (Selectable) Parameters Magnetic Field Strength RF Pulse timing Excitation pulse repetition (TR) Time to echo (TE) & inter-echo spacing (T ie ) Inversion (TI) RF pulse amplitude (flip angles) Gradient Amplitude & Timing b-value RF pulse excitation frequency & bandwidth Receiver bandwidth G.D. Clarke, AAPM 2004 1

Spin (Proton) Density Spin Echo Pulse Sequence Tissue Liver Kidney Muscle Heart % Water 70 % 75% 73% 78% Tissue Spleen Brain (GM) Brain (WM) Tendon % Water 78% 85% 75% 60% TX RX RF 1 (excitation) FID τ RF 2 (rephasing) τ Spin Echo From Akber SF, Physiol Chem Phys & Med 1996; 28: 205-238. TE Out of Phase Spins (Transverse Magnetization) Dephasing of MR Signal (Transverse Magnetization) Spins in Phase Out of Phase Precessing Off-resonance by 17,040 Hz 17, 040 Hz = 17, 040 cycles per second = 360 17, 040 degrees/sec 180 o (360 o 17, 040 degrees/sec) = 3 10-5 s = 30µs Coherent Signal Less Signal No Signal Signal decays exponentially: (after 90 o rf pulse) M xy = M o exp [- TE / T 2* ] where 1/T 2* = 1/T 2 + (γ/2π) * B (Magnetic field o inhomogeneity) Rephasing of the MR Signal (Spin Echoes) 4 3 1 2 Signal Dephasing 2 3 4 1 180 o pulse Rephased Signal A rephasing 180 o radio frequency pulse, changes the phase of the net magnetization vectors by 180 o. Thus spins precessing in the same direction come together of moving apart. This forms a SPIN ECHO. Longitudinal (T 1 ) Relaxation Longitudinal relaxation is a recovery process: After 90 o pulse - all spin energy is kinetic Spins lose energy to their environment, signal decreases After 5*T 1, full potential of net magnetization is reestablished G.D. Clarke, AAPM 2004 2

Comparison of T 1, T 2 and for Various Tissues Tissue T 1 (ms) T 2 (ms) Time Scale of MR Relaxation Processes Liver 675+142 54+8 Kidney 559+10 84+8 Muscle 1123+119 43+4 Gray Matter 1136+91 87+15 White Matter 889+30 86+1.5 Longitudinal Magnetization Transverse Magnetization M o exp (-TE/T 2* ) 1- exp (-TR / T 1 ) 100 s of ms Akber, 1996 (at 63 MHz) 10 s of ms Manipulating Contrast Rules of Thumb The weighting of image contrast Gradient Echo manipulating image contrast by varying the flip angle Spin Echo manipulating image contrast with 180 o refocusing pulses Inversion Recovery manipulating image contrast with 180 o inversion pulses TE controls T2 dependence TR controls T1 dependence T1 competes with T2 and proton density PD Weighted = long TR, short TE T1 Weighted = short TR, short TE T2 Weighted = long TR, long TE Pulse Sequence Classifications Spin-Echo Imaging Sequence rf 1 rf2 Name Gradient Echo Spin Echo Inversion Recovery RF Pulses One Two or more Three Contrast Weighting T1 or T2 * T1, PD or T2 T1 and T2 Application Fast imaging (3DFT) conventional Excludes certain tissues TX RX G sl G ro G pe Spin Echo G.D. Clarke, AAPM 2004 3

Spin Echo Imaging Effect of TR TR = 63 ms, NSA = 16 TR =125 ms, NSA = 8 TR = 250 ms, NSA = 4 TR = 500 ms, NSA = 2 Multi-Echo Acquisitions M o exp(-te/t 2 ) M o exp (-TE/T 2* ) Signal Strength 90 o 180 o 180 o SE2 180 o SE1 TE 2*TE 3*TE SE3 B o = 1.5 T FOV = 230 mm 256 x 256 st = 4 mm TE = 15 ms TR = 1 s, NSA = 1 TR = 2 s, NSA = 1 TR = 4 s, NSA = 1 Vlaardingerbroek & den Boer, 1999 G slice G read G phase Multiecho Image Matrix TR (s) 4 2 1.5 10 4 Sensitivity Plots (pulse sequence optimization) 0.9 M o Proton Density T1 T2 Most T 1 Weighted Most Proton Density Weighted 1.2 0.9 10 3 TR (ms) 10 2 0.1 M o Most T 2 Weighted TE 30 60 90 120 150 180 ms 0.6 10 10 10 2 10 3 10 10 2 10 3 10 10 2 10 3 TE (ms) TE (ms) TE (ms) Inversion Recovery Multi-Slice Imaging 180 o 90 o 180 o Spin Echo 180 o Inversion TI TR TE Excitation Refocusing Increased efficiency by scanning multiple planes during TR interval (Crooks L, Arakawa M, Hoenninger J et al. Radiology 1982; 143: 169-174.) +M o M = M o (1 2 exp[-ti/t 1 ]) -M o TI G.D. Clarke, AAPM 2004 4

STIR (Short TI Inversion Recovery) M z Fat 0 Liver Phase Compensated Inversion Recovery FSE TR =1400 TI = 100 TR =1400 TI = 280 TR =2000 TI = 280 TI Modulus Vlaardingerbroek & den Boer, 1999 FLAIR (FLuid Attenuated Inversion Recovery) Multiple Sclerosis Proton Density T2-Weighted FLAIR M z Brain M xy Lesion Lesion 0 CSF TI Brain T eff TR = 2350 TE = 30 TR=2350 TE= 80 Rydberg JN et al. Radiology 1994; 193:173-180 TR=2600 TE=145 TI= 11,000 TX RX G sl G ro G pe Steady-State Free Precession (SSFP or True FISP) rf Slice Select Dephasing Rephasing Read Out Phase Encode Field Echo Digitizer On Phase Rewinder M y /M o 0.40 0.20 0 Relative Sensitivity of Gradient Echo Sequences SSFP FISP FLASH T2 =200 ms M y /M o 0.40 0.20 0 SSFP FISP FLASH T2=30 ms Also called FAST, ROAST 0 30o 60o 90o 120o 150o 180o Flip Angle 0 30o 60o 90o 120o 150o 180o Flip Angle G.D. Clarke, AAPM 2004 5

Gradient-Echo Imaging Cystic Mass Spin Echo TR/TE = 2500/90 NSA =1 5 mm FISP TR/TE = 22/10 NSA =8 8 mm True FISP TR/TE = 16/6 flip = 40 o NSA =8 5 mm True FISP TR/TE = 16/6 flip = 70 o NSA =8 5 mm Reference: Wehrli, Fast-Scan Magnetic Resonance. Principles and Applications Signal-to-Noise Ratio Signal-to-noise in an RF coil: v ωom o B ( r ) dv S 1 N 4kT BW R t SNR and Imaging Parameters FOV SNR M ro ro FOV M ph ph z NSA BW rx S = signal N = noise ω o = resonant frequency M o = net magnetization B 1 (r) = RF field distribution dv = elemental volume of tissue k = Boltzman s constant T = patient temperature BW = receiver bandwidth R t = total resistance (coil + patient) FOV = field of view M = matrix size n = number of signals averaged BW rx = receiver bandwidth ro = read out (frequency encoding) direction z = slice thickness ph = phase encoding direction Matched Bandwidth Matched Bandwidth Bandwidth on ADC 1 is limited by finite T max & desire to minimize TE 1. T 1 weighting and SNR is sacrificed more by T 2 decay than by BW T 2 -weighted image - lots of for ADC 2, poorer SNR. It would be nice to decrease BW (increase T max ) ADC 90 o 180 o TE1 20 ms TE2 60-80 ms ADC t Mugler & Brookeman, Magn Reson Imag 1989; 7:487-493. Mugler & Brookeman, Magn Reson Imag 1989; 7:487-493. G.D. Clarke, AAPM 2004 6

T1W Matched Bandwidth = Unmatched Distortion T2W Smaller BW on T2W image leads to increased image distortion Chemical Shift Chemical shift (fat-water) ~3.5 ppm At 1.5T: 6 42.6MHz f fat water = 3.5 10 1.5T 220 Hz T At 0.5T: 6 42.6MHz f fat water = 3.5 10 0.5T 73 Hz T field strength. chemical shift Chemical Shift Artifact Occurs in Readout direction Conventional SE Phase encode direction Echo-Planar Controlled by Fat Pre- Saturation STIR sequence Chemical shift artifact seen in axial image of kidney Contrast-to-Noise Ratio CNR = n ( ) I a I b σ σ = standard deviation of the background signal (the noise) n = number of signals acquired I a,i b = signal intensities from tissues a and b Steady state free precession (SSFP) Signal is produced by rapidly applying balanced gradients and RF pulses. Use to maximize the overall signal. FIESTA (fast imaging employing steady state acquisition) Balanced FFE (fast field echo) True FISP (fast imaging with steady precession) IR Turbo-FLASH 180 o TI α α α α α α α FID Train (one for each image line) TR Hinton et al. Invest Radiol 2003; 38:436 G.D. Clarke, AAPM 2004 7

3D FLASH vs. 3D Turbo FLASH FSE Pulse Sequence RF G ro G pe Overall Signal Echo T ie IR Turbo FLASH FLASH TI = 500 ms, TR/TE = 10/4.5 TR/TE = 12/4.5 NSA =1, flip = 10 o NSA = 1, flip = 8 o Center of K-space ETL = Echo Train Length Things That have Blurry point spread functions are Sharp in the Image FSE Point Spread Function Depends on T 2 K-space FT Image Space Effect of Echo Spacing SE FSE FSE Fat Sat Multi-Shot EPI Pulse Sequences Echo 1 Echo 2 Echo 3 Echo 4 Spoiler ES = 60 ms, TF =3 ES = 30 ms, TF =7 FSE FSE FSE Slice Select Blip 2 Blip 3 Blip 4 Spoiler Start Blip Rewind ES = 15 ms, TF =15 ES = 13 ms, TF =20 ES = 10 ms, TF =27 Signal-to-noise decreases for short T2 tissues (gray & white matter) leading to a decrease in spatial resolution Vlardengerbrook & den Boer, 1999 Preparatory (Fat Sat) RF Dephase/ Excitation Rephase t s T ie TR T tot G.D. Clarke, AAPM 2004 8

Comparison of EPI and SE Magnetization Transfer PROTON SPECTRUM 0 Frequency (Hertz) Free Water Lipids Bound Water T2-weighted SE Single shot EPI Nine-shot EPI TR 2400, TE 80 TE =80 TR 3000, TE 21 192x256; 7:44 128x128; 120 ms 252x256; 1:06 0 217 Hz 1500 Hz Frequency (Hertz) Gradient Echo with MTC Pulse Magnetization Transfer Contrast TX RX G sl G ro G pe Spoilers Off-resonance rf pulse RF excitation Digitizer On Slice Select Phase Encode Field Echo Rephasing Read Out Dephasing Crushers or Spoilers Multislice FSE: Magnetization Transfer Contrast Enhances T 2 - Weighted Appearance Slice thickness - RF bandwidth Crafted RF Pulses thinner slices require longer RF pulses z ω FT sin x A sinc function ( x ) envelope on the r.f. pulse produces a nearly square excitation profile of the phantom.. FT BW Bandwidth = 1 period FT t p frequency G.D. Clarke, AAPM 2004 9

MRI Slice Thickness Cranial Nerves Signal ramps have a slope of 10:1 Signal from ramp is 10 x slice thickness Two ramps are used to compensate for inplane rotation of the phantom Phantom does not compensate for tilting backwards or swaying left-right 10 mm slice 3 mm slice Frequency Encoding & FOV The field of view (FOV x ) is determined by the range of frequencies in the x- axis which are sampled Bandwidth FOVx = γ G The receiver bandwidth is not the same as the RF bandwidth BW FOV x G x FOV x FOV must be sufficient to cover all tissue in the slice to avoid aliasing x G Phase Encode Gradient The maximum phase-encoding encoding gradient is : ph max 2π = γ BW FOV π N = γ FOV T where FOV is the desired field of view The phase-encoding encoding gradient increment is : G ph π 1 = γ FOV T Note that a factor of two is gone from G ph-max because G ph runs from - G ph=max to G ph=max y y Sampling the Signal CHESS Each signal is digitally sampled to produce one line in the k- space matrix RF 90 0 sat 90 0 180 0 k = γ G t x x MHz mt kx = msec T m cycles = m 1 kx = FOV x x G spoiler Signal This is typically accomplished by preceding a SE or FSE sequence with a 90 o pulse that is frequency, not spatially, selective. G.D. Clarke, AAPM 2004 10

Fat Suppression Spatial Saturation Blood Flow 90 o Sat Pulse 90 o -180 o SE Pulses T1W images without (left) and with (right) fat suppression. Spatial Saturation a. No Sat Pulse b. Inferior Sat Pulse c. Superior Sat Pulse d. Inferior and Superior Sat Pulses GE Signa Applications Guide, Vol 1 Gadolinium Contrast Agents GD-DTPA, Gd-DOTA, Gd-HP-DO3A: the old standards, generally diffusable but not in brain due to Blood Brain Barrier In low doses decrease T1 In higher doses decrease T1 and T * 2 NEW INTRAVASCULAR AGENTS - Blood pool agents Magnetic iron oxide particles Bind serum albumin, have higher relaxivity Gadolinium Contrast Agents T 1 - weighted Images Normal With Gd Contrast Dynamic Contrast Studies Meningioma Dynamic study of breast tumor Curves denote difference between uptake of normal glandular tissue & lesion Vlaardingerbroek & den Boer, 1999 G.D. Clarke, AAPM 2004 11

Attenuation Due to Diffusion 2 2 2 A( TE) = A(0) exp[ γ G Dappδ α ( )] 4 2 δ Where: α=π/2; G is amplitude of diffusion sensitive gradient pulse; δ is duration of diffusion sensitive gradient; is between diffusion sensitive gradient pulses; D app is the apparent diffusion coefficient Diffusion Weighted Imaging: Prototype Pulse Sequence 90 o G δ 180 o Stejskal EO & Tanner JE, 1965. 42: 288-292 δ G 2 2 b = γ G δ ( ) 2 δ 3 The b-value Diffusion EPI Pulse Sequence Controls amount of diffusion weighting in image RF 90 o SS 180 o The greater the b-value the greater the area under the diffusion-weighted gradient pulses FE PE G x G x longer TE stronger and faster ramping the gradients MR signal ~100 ms Diffusion Imaging of Infarcts Biorelaxometry SE Echo Planar TE = 80 ms SE Echo Planar b = 1205 s mm -1 Koenig, et. al, 1983 G.D. Clarke, AAPM 2004 12

Specific Absorption Rate (SAR) The patient is in an RF magnetic field that causes spin excitation (the B1 field) The RF field can induce small currents in the electrically conductive patient which result in energy being absorbed. The RF power absorbed by the body is called the specific absorption rate (SAR) SAR has units of watts absorbed per kg of patient If the SAR exceeds the thermal regulation capacity the patient s body temperature will rise. Scan Parameters Effecting SAR Patient size: SAR increases as the patient size increases directly related to patient radius Resonant frequency: SAR increases with the square of the Larmor frequency (ω o ) RF pulse flip angle: SAR increases as the square of the flip angle (α Number of RF pulses: SAR increases with the number of RF pulses in a given SAR Effects on Pulse Sequences Decrease number of slices per study Requires decreased flip angles, even in gradient echo sequences Forces increases in TR 1.5T 3 T Uniformity at 3 Tesla Limits use of Fast Spin Echo imaging B1 field maps in a conductive saline phantom (18 cm diameter) RL Greenman et al. JMRI 2003, 17(6): 648-655 655 Summary RF pulses can be crafted to select slices RF pulse timing (TE, TR & TI) is used to manipulate soft tissue contrast in MR images Gradient echo imaging with partial flip angles is faster than spin echo imaging Spin Echoes correct for poor magnetic fields Inversion recovery can be used to get rid of fat or CSF from images Contrast agents are used to manipulate tissue T 1 Summary Presaturation schemes can effect flow, chemical shift and T1-weighted contrast Fast imaging methods are often applied to imaging physiology as well as anatomy Typically, SNR is reduced with increases in resolution, contrast and imaging speed Various image parameters will have to be adjusted for each magnetic field strength G.D. Clarke, AAPM 2004 13