Quality Control and the ACR’s MRI Accreditation Program

Geoffrey D. Clarke, Ph.D.University of Texas Health Science Center

at San Antonio

Radiological Sciences Division

Overview

• ACR MRI Accreditation Program

• MRI Quality Control Program

– Definitions

– Technologist’s Daily & Weekly QC Tasks

– Physicist’s Responsibilities

– Annual Tests for the Physicist

ACR MRI Accreditation ProgramFeatures

• Evaluates effectiveness of quality control measures

• Will collect findings to further the development of quality control information

• Qualified Medical Physicist should be responsible for overseeing the equipment quality control program

ACR MRI Accreditation ProgramMRI Survey Agreement

• Official request for ACR Accreditation

• Site agrees to provide all documentation, including but not limited to quality control logs, films, records, or any necessary information requested by the survey team

• Agree to use the ACR MRI phantom

• Tests run in a short time

• Pulse sequences as compatible as possible with all commercial MRI scanners

• Specific measurements to account for :– Geometric Distortion

– Slice thickness & Position

– Factors Affecting Image SNR (resolution, bandwidth, ghosting)

– Image Uniformity

ACR PhantomACR Phantom• Affordable

($730)• Results

easily evaluated

• Affordable ($730)

• Results easily evaluated

Documents for Using Phantom

ACR MRI Accreditation ProgramQuality Control Section

• Refers to tests put forth in the ACR Standard for MRI

• All tests to be carried out in accordance with written procedures and methods

• Preventative maintenance:

– Documented by qualified service engineer

– Repairs documented and records maintained by the MR site

ACRMagnetic

ResonanceImaging QualityControl Manual2001

Radiologist’s Responsibilities Eleven specific responsibilities including: • To ensure that an effective quality control program

exists for all MRI• To select the technologist to be the primary quality

control technologist• To ensure that appropriate test equipment and

materials are available to perform the technologist's QC tests.

• To arrange staffing and scheduling so that quality control tests can be carried out.

• To select a qualified medical physicist or MRI scientist To ensure that records are properly maintained and updated in the MRI QC procedures manual.

MRI QC Technologist’s Responsibilities

• Weekly magnetic resonance image quality control procedures (daily recommended)

• Weekly quality control of hard copy and soft copy Images

• Routine visual inspection of equipment

Weekly Technologist’s Tests• Center frequency• Table positioning• Setup & Scanning

• Geometric accuracy• High contrast resolution• Low contrast detectability• Artifact analysis• Film quality control• Visual Checklist

Technologist QC

• Daily (weekly) tests

• Cheat sheets

• Review data after ten days to establish baseline values and variability

• Daily (weekly) tests

• Cheat sheets

• Review data after ten days to establish baseline values and variability

Measurement of Central Frequency

• Can be performed by technologist

• Record center frequency value on ACR phantom or manufacturer’s phantom

• Automated analysis and recording often available on modern MRI systems

Central FrequencyCentral Frequency

•Specific but not sensitive

•Action Criteria:Change in Hz from previous day > 2 * resonant frequency in MHz

• Suggestive of:Magnet driftRF instability

•Specific but not sensitive

•Action Criteria:Change in Hz from previous day > 2 * resonant frequency in MHz

• Suggestive of:Magnet driftRF instability

Transmitter Gain/Attenuation

• Reflects power required to optimize RF pulse:

– Depends on coil, phantom, pulse sequence, etc.

– Should remain constant over time if nothing in pulse sequence or hardware has changed

Terminology & Units Vary with Vendor• GE: displayed on screen (dB)• Philips: under system performance

parameters– rf_act_drivescale

• Siemens: “options” – “adjustments”– Frequency– Transmitter amplitude (temp) (V)

• Toshiba: “acquisition window”– Center frequency (MHz)– RF level

Geometric Accuracy

• Measure distance along main axes of phantom

• Compare with known values

Potential Causes of Geometric Accuracy Failures

1. Phantom mispositioning

2. Gradient miscalibration

3. Bo inhomogeneity

a. Ferromagnetic objects in magnet

b. Poor magnet shimming

4. Gradient non-linearity

5. Inappropriate receiver bandwidth

6. Poor eddy current compensation

7. Combination of two or more of above

Geometric Accuracy - Axial

Failure Due To: • Miscalibrated Gradients • Low Receiver Bandwidth

• High Bo Inhomogeneities

BW = 7.4 kHz BW = 3.6 kHz

Slice #1Slice #1

Gradient Correction

• If gradients are inherently non-linear gradient correction may be applied

Percent Geometric Distortion T1-Weighted, Central Slice

0.00%

2.00%

4.00%

6.00%

8.00%

10.00%

1-Oct

31-Oct

30-Nov

30-Dec

29-Jan

28-Feb

29-Mar

28-Apr

28-May

27-Jun

27-Jul

26-Aug

25-Sep

25-Oct

Date of Measurement

System #1 System #2 System #3

Poor Gradient Calibration

Poor Gradient Calibration

Assessment of MR Image Quality

• Performed by Technologist

• Using ACR Phantom

• High Contrast Resolution

• Low Contrast Detectability

RF Noise/Leaks/ Spikes

Single frequency

artifact shows up as

zipper in middle of image.

0.90 mm0.90 mm1.0 mm1.0 mm1.1 mm1.1 mm

ChemShiftChemShift

Slice ThicknessSlice Thickness

Resolution InsertResolution InsertResolution InsertResolution Insert

ACR MRI PhantomACR MRI PhantomResolution InsertResolution Insert

ZOOM = 2ZOOM = 2

Spatial Resolution Matrix: Spatial Resolution Matrix: Registration with PhantomRegistration with PhantomSpatial Resolution Matrix: Spatial Resolution Matrix: Registration with PhantomRegistration with Phantom

ImageMatrixImageMatrix

ResolutionResolutionHolesHolesResolutionResolutionHolesHoles

PARTIAL VOLUME ARTIFACT

High Contrast Spatial Resolution

• Evaluate conspicuity of holes arranged in two square arrays

• Has to avoid partial-volume artifact…

UL

LR

High Contrast ResolutionHigh Contrast Resolution

•Specific but not sensitive

•Action Criteria:Any reduction in # of holes seen

• Suggestive of:Increased eddy currentsPoor gradient calibration

Poor Bo uniformity

Reduced stability of system

•Specific but not sensitive

•Action Criteria:Any reduction in # of holes seen

• Suggestive of:Increased eddy currentsPoor gradient calibration

Poor Bo uniformity

Reduced stability of system

Slice #10Slice #10 Slice #12Slice #12

Low Contrast DetectabilityLow Contrast Detectability

5.1% Contrast Detectability5.1% Contrast Detectability3.7% Contrast Detectability3.7% Contrast Detectability

Low Contrast Detectability1.5 T System: T1-Weighted Scan

0.00

5.00

10.00

15.00

20.00

25.00

1-May 31-May 30-Jun 30-Jul 29-Aug 28-Sep 28-Oct 27-Nov 27-Dec 26-Jan 25-Feb 27-Mar 26-Apr

Date

Ho

le S

ize

Vis

ual

ized

(mm

)

1.4% contrast

2.4% contrast

3.7% contrast

5.1% contrast

LCD and Signal-to-NoiseT

ota

l Nu

mb

er o

f S

po

kes

To

tal N

um

ber

of

Sp

oke

sT

ota

l Nu

mb

er o

f S

po

kes

To

tal N

um

ber

of

Sp

oke

s

Signal-to-Noise RatioSignal-to-Noise RatioSignal-to-Noise RatioSignal-to-Noise Ratio0 50 100 150 200 250 300 3500 50 100 150 200 250 300 3500 50 100 150 200 250 300 3500 50 100 150 200 250 300 350

4040

3535

3030

2525

2020

1515

1010

55

00

4040

3535

3030

2525

2020

1515

1010

55

00

Low Contrast DetectabilityLow Contrast Detectability

•Sensitive but not specific•Action Criteria:

•Slice used dependent on Bo field strength•Sustained 5 row decrease in number of hole sets seen

• Suggestive of:•Reduced stability of system

•Sensitive but not specific•Action Criteria:

•Slice used dependent on Bo field strength•Sustained 5 row decrease in number of hole sets seen

• Suggestive of:•Reduced stability of system

ACR MRI StandardHardcopy Image QC Tests

• Sensitometric Measurements for Film Processors

• Hard Copy of SMPTE test pattern

• Similar to process used for mammography program

• Laser camera film less sensitive to temperature changes

Laser Film QCWeekly:Weekly:

View SMPTE patternView SMPTE pattern

Verify gray levelsVerify gray levels

• 0/5% & 95/100% 0/5% & 95/100% patchespatches

Film 6 on 1Film 6 on 1

• 4 on 1 if necessary4 on 1 if necessary

Plot OD of Plot OD of

• 10%, 40% & 90& 10%, 40% & 90& patchespatches

Observe film for Observe film for artifactsartifacts

Weekly:Weekly:

View SMPTE patternView SMPTE pattern

Verify gray levelsVerify gray levels

• 0/5% & 95/100% 0/5% & 95/100% patchespatches

Film 6 on 1Film 6 on 1

• 4 on 1 if necessary4 on 1 if necessary

Plot OD of Plot OD of

• 10%, 40% & 90& 10%, 40% & 90& patchespatches

Observe film for Observe film for artifactsartifacts

SMPTE Test PatternSMPTE Test Pattern

Action Limits

SMPTE patch

OD Control Limit

0 2.45 0.15

10% 2.10 0.15

40% 1.15 0.15

90% 0.30 0.08

Documenting the Quality Control Program

• Data form for daily equipment quality control – ACR MRI QC manual, pg. 64

• MRI Facility quality control visual checklist– ACR MRI QC manual, pg. 65

• Laser film printer control chart– ACR MRI QC manual, pg. 66

If QC Test Fails

• Common errors-– Check for magnetic objects in bore

– Re-seat head coil

– Reposition & landmark phantom

– Make sure scan room door securely closed

• Repeat daily QC scan procedures

• Record results again in QC notebook

Responsibilities of the Qualified Medical Physicist

or MRI Scientist • Write Purchase Specifications

• Perform Acceptance Testing – Baseline Measurements

• Determine Action Limits

• Set up Daily/Weekly QC Tests

• MRI equipment performance review

Annual Survey Tests• Magnetic Field Homogeneity

• Slice Position Accuracy

• Slice Thickness Accuracy

• Radio Frequency Coil Checks

– Volume coils

1. Signal-to-noise ratio

2. Percent integral uniformity

3. Percent signal ghosting

– Surface Coils SNR Tests

• Inter-Slice RF Interference

• Soft Copy Displays (monitors)

Magnetic Field HomogeneityMagnetic Field Homogeneity

oo oo oo

Fourier transform of signal produces a Lorentzian peak in

well-shimmed magnet

Fourier transform of signal produces a Lorentzian peak in

well-shimmed magnet

Magnet firld homogenity can be characterized using FWHM of resonance

peak

Magnet firld homogenity can be characterized using FWHM of resonance

peak

Denotes a totally uniform magnetic

field.All signal is at

resonant

frequency, o.

Denotes a totally uniform magnetic

field.All signal is at

resonant

frequency, o.

Ideal HomogeneityIdeal Homogeneity Good HomogeneityGood Homogeneity Poor HomogeneityPoor Homogeneity

FWHMFWHM

FWHMFWHM

Magnetic Field Homogeneity

Phase images from GRE sequences with 13.9 ms difference in TE’s

Phase and Unwrapped

Phase Images

John Pauley, Stanford Univ.John Pauley, Stanford Univ.

Magnetic Field Homogeneity

• Overall, the phase mapping technique provides the best mechanism for evaluating field homogeneity.

• For some system service personnel may have to provide use of phase-mapping acquisition and analysis tools.

• Filmed copy of vendor’s final homogeneity map and shim coefficients is useful for documentation and establishing a baseline.

Slice Position Accuracy

• Uses Crossed-Wedges as Reference for Positioning and Slice Spacing Accuracy

• MRAP pass criterion: magnitude of bar length difference 5 mm.

• The actual displacement is ½ of the measured difference

• ACR Accreditation criteria are very weak, physicist may want to hold manufacturer to a higher standard

Slice Position Accuracy

SLICE #1SLICE #1SLICE #1SLICE #1 SLICE #11SLICE #11SLICE #11SLICE #11

Slice PositionSlice Position Slice SpacingSlice Spacing

Crossed wedges should be of equal lengthCrossed wedges should be of equal length if position and spacing are accurateif position and spacing are accurate

Crossed wedges should be of equal lengthCrossed wedges should be of equal length if position and spacing are accurateif position and spacing are accurate

Slice Position Accuracy

• Causes of poor performance:

• Operator error

• Table positioning shift

• Miscalibrated gradients

• High Bo inhomogeneities

MRI Slice Thickness

• Signal ramps have a slope of 10:1

• Signal from ramp is 10 x slice thickness

• Two ramps are used to compensate for in-plane rotation of the phantom

• Phantom does not compensate for tilting backwards or swaying left-right

Slice Thickness Measurement

1. Use slice 1 of ACR T1.

2. Magnify by 2 to 4. Adjust window/level to see signal ramps. (Set window at minimum.)

3. Use rectangular ROI to measure mean of middle of each signal ramp. Take average.

Slice Thickness– Lower display to

one/half the average ramp signal.

– Measure lengths of top and bottom ramps and calculate slice thickness.

bottom top

bottom top 0.2 Thickness Slice

Slice Thickness

• ACR-MRAP: slice thickness measured should be + 0.7 mm of prescribed value

– + 14% error on 5mm slice, may be too generous

• Corrective actions:

– Check Axial Site Series Images

– Replace cables & connectors, look for other sources of distorted RF pulse shape in RF electronics

– Try switching RF coils

– Check gradient calibration

MRI Equipment Performance Evaluation Site: _____________________________ Date: ________ MRAP Number: ____________________ Serial Number: ___________ Equipment: MRI System Manuafacturer: _________________ Model : ________

Processor Manufacturer: _________________ Model: _________

PACS Manufacturer: _________________ Model: _________

ACR MRAP Phantom Number used: _________

1. Magnetic Field Homogeneity Method Used (check one): Spectral Peak ___ Phase Difference ___ Other (describe) __________________________ Measured Homogeneity: Diameter of Spherical Homogeneity Volume (cm) (ppm) ________ _______ ________ _______ ________ _______

2. Slice Position Accuracy From Slice Positionss #1 and #11 of the ACR Phantom: Wedge (mm)

Slice Location #1 ________

Slice Location #11 ________

3. Slice Thickness Accuracy From Slice Position #1 of the ACR Phantom: Slice Thickness Top ______ Calculated slice

(fwhm in mm) Thickness (mm) ______

Bottom ______

Duplicate these forms so they will be available for repeated use.

=+ = -

=+ =

-

Site & Equipment

Data

Bo

Homogeneity

Slice PositionAccuracy

Slice ThicknessAccuracy

Bird-Cage Type Head Coil

RF coils produce uniformity patterns characteristic of their design.RF coils produce uniformity patterns characteristic of their design.

Uniformity Pattern

Birdcage CoilHigh Field

Birdcage CoilHigh Field

Solenoid CoilLow Field

Solenoid CoilLow Field

ACR PhantomSlice #7

Volume RF Coil MeasurementsVolume RF Coil Measurements

Image Intensity Uniformity

• Performance criteria: PIU 90%

• Measurement Considerations:• Display may not show signal values• Display may not allow user to set signal

display level• There may not be a well-defined

high/low intensity level

percent integral uniformity = 100

1

( )

( )

high low

high low

Image Intensity Uniformity

• Causes of failure:

• Poor phantom centering in head coil (usually AP)

• Ghosting

• Motion or vibration

• Mechanical failure in head coil

Phased- Array Coils•The signal is viewed from more “angles”•Using four channels does not produce 4 * S/N

Abdomen-pelvisphased-array

Cervical-Cranial Phased Array

TorsoPhased Array

General Features of MRI Phantoms

– Nonsignal-producing container

– Proton density similar to water

– Shorten T1: NiCl & CuSO4

– Mimic Conductivity of tissues: NaCl

Agarose Agarose Gel Gel

Phantom Phantom with Insertswith Inserts

Uniform Uniform Spherical Spherical PhantomsPhantoms

Phantoms Developed by Users

Surface RF Coil Measurements

Volume Coil Data% Image Uniformity Max Signal

Min Signal

Signal-to-Noise Mean Signal

SD of Background Signal

Percent Signal Ghosting

Ghost Signal

Mean Signal

Background Signal

Surface Coil DataMaximum Signal-to-Noise

Maximum signal

SD of Background Signal

-

Volume CoilData Recorded

Calculated Values:Uniformity

SNRGhosting

Surface Coil

Signal-to-Noise

4. RF Coil Performance Evaluation

A. VOLUME RF COIL - RF Coil Description: __________________________ Date: ____________

Phantom Description: ___________________________________________

Pulse Sequence: Type: ____ TR: _____ TE: ______ flip angle _____ degrees

FOV: _____ cm2 Matrix: ___________ BW: _________kHz ; NSA ___

Slice thickness ______mm; spacing _______ mm

TX attenuation (or gain) __________ Data Collected:

Mean Signal

Maximum Signal

Minimum Signal

Background Signal

Noise Standard Deviation

Ghost Signal

Calculated Values:

Signal-to-Noise Ratio

Percent Image Uniformity

Percent Signal Ghosting

B. RF SURFACE COIL -

RF Coil Description: __________________________ Date: ____________

Phantom Description: ___________________________________________

Pulse Sequence: Type: ____ TR: _____ TE: ______ FOV: _____ cm2

Matrix: ___________ BW: _________kHz ; NSA ___

Slice thickness ______mm; spacing _______ mm

TX attenuation (or gain) __________

Maximum Signal

Noise Standard Deviation

Maximum Signal-to-Noise Ratio

Image uniformity distribution OK? ________

Image ghosting OK? ________

HARD COPY IMAGE: Window width ________ Window level _______

Several copies of this page may be required to report on all RF coils.

ACR MRI QC Manual, pg. 125ACR MRI QC Manual, pg. 125

Slice Cross-Talk Measurements

1. Position 5mm slices on the uniform volume

2. Repeat measurements decreasing the slice gap :

Series # 1 2 3 4

No. Slices 11 11 11 11

Slice Gap (mm) min 0.5 1.0 5

3. Measure the signal-to-noise ratio (SNR) for each of the image sets.

4. Plot the SNR vs. percentage slice gap

Signal-to-Noise vs. Inter-slice Gap

SE, TE = 20FE, TE =8TFE, TE =4

100%

90%

80%

70%0% 25% 50% 75% 100%

Percentage Slice Gap

Sig

na

l-to

-No

ise

Rat

io (

per

cen

t of M

axim

um

)

Soft Copy Displays

• Requires precision luminance meter

• Four tests– Maximum and minimum

luminance – Luminance uniformity – Resolution – Spatial Accuracy

-

RF Slice Interference

Soft CopyDisplays

Review of Routine QC

Program

5 . I n t e r s l i c e R F I n t e r f e r e n c e

P h a n t o m D e s c r i p t i o n : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

P u l s e S e q u e n c e : T y p e : _ _ _ _ T R : _ _ _ _ _ T E : _ _ _ _ _ _ F O V : _ _ _ _ _ c m 2

M a t r i x : _ _ _ _ _ _ _ _ _ _ _ B W : _ _ _ _ _ _ _ _ _ k H z ; N S A _ _ _

N u m b e r o f s l i c e s _ _ _ _ _ _

6 . S o f t C o p y D i s p l a y s

M o n i t o r D e s c r i p t i o n : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

M a x i m u m L u m i n a n c e : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ C d m - 2 .

M i n i m u m L u m i n a n c e : _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ C d m - 2 .

L u m i n a n c e U n i f o r m i t y :

A v e r a g e o f v a l u e s o b t a i n e d i n f o u r c o r n e r s o f s c r e e n : _ _ _ _ _ _ C d m - 2 .

L u m i n a n c e m e a s u r e d i n c e n t e r o f s c r e e n : _ _ _ _ _ _ C d m - 2 .

P e r c e n t d i f f e r e n c e : _ _ _ _ _ _ _ _ %

| ( C e n t e r – A v e r a g e C o r n e r s ) / ( C e n t e r ) x 1 0 0 % < 3 0 % |

7 . E v a l u a t i o n o f S i t e ’ s T e c h n o l o g i s t Q C P r o g r a m

4 ) S e t u p a n d p o s i t i o n i n g a c c u r a c y : ( d a i l y ) _ _ _ _ _ _ _ _ _

5 ) C e n t e r F r e q u e n c y : ( d a i l y ) _ _ _ _ _ _ _ _ _

6 ) T r a n s m i t t e r A t t e n u a t i o n o r G a i n : ( d a i l y ) _ _ _ _ _ _ _ _ _

7 ) G e o m e t r i c A c c u r a c y M e a s u r e m e n t s : ( d a i l y ) _ _ _ _ _ _ _ _ _

8 ) S p a t i a l R e s o l u t i o n M e a s u r e m e n t s : ( d a i l y ) _ _ _ _ _ _ _ _ _

9 ) L o w C o n t r a s t D e t e c t a b i l i t y : ( d a i l y ) _ _ _ _ _ _ _ _ _

1 0 ) F i l m Q u a l i t y C o n t r o l ( w e e k l y ) _ _ _ _ _ _ _ _ _

V i s u a l C h e c k l i s t : ( w e e k l y ) _ _ _ _ _ _ _ _ _

Series Slice Signal-Number Gap to-Noise

(mm) Ratio

1

2

Measure

d SNR

100%

90%

80%

70%0% 25% 50% 75% 100%

Inter-slice Gap (percent of slice thickness)

3

4

Series Slice Signal-Number Gap to-Noise

(mm) Ratio

1

2

Measure

d SNR

100%

90%

80%

70%0% 25% 50% 75% 100%

Inter-slice Gap (percent of slice thickness)

3

4

ACR MRI QC Manual, pg. 127ACR MRI QC Manual, pg. 127

MRI Equipment Evaluation Summary

Site ___________________ Report Date: __________ System MRAP #_____________ Survey Date: __________

MRI System Manufacturer ___________ Model: __________ Physicist/MRI Scientist: ____________________ Signature: ________________________________

Equipment Evaluation Tests

Pass / Fail 1. Magnetic Field Homogeneity: _________ 2. Slice Position Accuracy _________

3. Slice Thickness Accuracy _________ 4. RF Coils’ Performance

a. Volume Coils’ Signal-to-Noise Ratio _________ b. Volume Coils’ Image Uniformity _________ c. Volume Coils’ Ghosting Ratios _________ d. Surface Coils’ Signal-to-Noise Ratio _________ 5. Inter-slice RF Interference _________ 6. Soft copy displays _________

Medical Physicist’s or MRI Scientist’s Recommendations for Quality Improvement:

______________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

SummarySummarySheetSheet(pg. 131)(pg. 131)

Successful MRI QC ProgramSuccessful MRI QC Program

1.1. Technologists run QC scans on a Technologists run QC scans on a daily basisdaily basis

2.2. If exceed action criteria – repeat QC If exceed action criteria – repeat QC procedureprocedure

3.3. Physicist reviews QC data annuallyPhysicist reviews QC data annually

4.4. Record data - report problems to Record data - report problems to service service

5.5. Have service record problems and Have service record problems and solutions in a service logsolutions in a service log

1.1. Technologists run QC scans on a Technologists run QC scans on a daily basisdaily basis

2.2. If exceed action criteria – repeat QC If exceed action criteria – repeat QC procedureprocedure

3.3. Physicist reviews QC data annuallyPhysicist reviews QC data annually

4.4. Record data - report problems to Record data - report problems to service service

5.5. Have service record problems and Have service record problems and solutions in a service logsolutions in a service log