1

PET Scanner Performance: Quality Assurance and

Acceptance Testing

PET Scanner Performance: Quality Assurance and

Acceptance Testing

Brad Kemp, PhDBrad Kemp, PhD

Department of RadiologyDepartment of Radiology

Mayo ClinicMayo Clinic

TU-A-I-617-2

IntroductionIntroduction

Quality ControlQuality ControlProcedures required to ensure that the distribution of radiationProcedures required to ensure that the distribution of radiation

emitted from a patient is accurately reflected in the measured remitted from a patient is accurately reflected in the measured raw aw data.data.

FacetsFacets•• Quality assurance:Quality assurance:

–– verify scanner is operating properlyverify scanner is operating properly

–– identify problems prior to scanning patientsidentify problems prior to scanning patients

•• Acquisition of corrections to compensate for known imperfectionsAcquisition of corrections to compensate for known imperfectionsin data measurementin data measurement

Quality AssuranceQuality Assurance

Quality Assurance Test RequirementsQuality Assurance Test Requirements

•• Well defined regimen of measurementsWell defined regimen of measurements

–– Quick and easy to conductQuick and easy to conduct

–– Sensitive to modes of failure of the scannerSensitive to modes of failure of the scanner

–– Preferably quantitative, not qualitativePreferably quantitative, not qualitative

Quality Assurance DataQuality Assurance Data

•• Blank Scan Sinograms: On standBlank Scan Sinograms: On stand--alone PET systems, used with alone PET systems, used with transmission sinograms to create attenuation correction factorstransmission sinograms to create attenuation correction factors

•• Emission Sinograms: Scans of uniform cylinderEmission Sinograms: Scans of uniform cylinder

Bad blockGood blank

Sinogram of a Uniform Cylinder

Blank Scan Sinograms

2

Quantitative Analysis of QA DataQuantitative Analysis of QA Data

•• Quantitative analysis is an important aspect of the QA of a Quantitative analysis is an important aspect of the QA of a PET scannerPET scanner

•• Quantitative analysis of QA data: Quantitative analysis of QA data:

-- validates system calibrationvalidates system calibration

-- provides information on crystal, block or module (bucket)provides information on crystal, block or module (bucket)efficiencies with respect to system averageefficiencies with respect to system average

-- compare to baseline data compare to baseline data -- monitor system stabilitymonitor system stability

-- ensures validity of normalizationensures validity of normalization

Quality AssuranceQuality Assurance

Detector and Electronics CharacterizationDetector and Electronics Characterization–– Singles mode detector calibrationSingles mode detector calibration

•• Crystal mapCrystal map•• PMT gain adjustmentPMT gain adjustment•• Energy mapEnergy map

–– Coincidence timing calibrationCoincidence timing calibration

Did the engineer calibrate the scanner properly?Did the engineer calibrate the scanner properly?

Crystal MapCrystal Map

Purpose: Purpose: •• Map the position of the detected event to a specific crystalMap the position of the detected event to a specific crystal•• Obtained quarterly or after detector maintenanceObtained quarterly or after detector maintenance

Count distribution in a blockCount distribution in a block Crystal Map Look Up TableCrystal Map Look Up Table

QA of crystal maps created during calibration of the scannerQA of crystal maps created during calibration of the scanner

Verification of Crystal MapVerification of Crystal Map

3

Purpose:Purpose:•• Balances the gain characteristics of the Balances the gain characteristics of the PMTsPMTs in a blockin a block•• Compensates for PMT gain drift with temperature, age, Compensates for PMT gain drift with temperature, age, ……•• Acquired weekly (daily) or after detector maintenanceAcquired weekly (daily) or after detector maintenance

PMT Gain AdjustmentPMT Gain Adjustment

-6.0

-5.0

-4.0

-3.0

-2.0

-1.0

0.0

28-Feb 7-Mar 14-Mar 21-Mar 28-Mar

Date

Blo

ck B

ias

(%)

Morning

Afternoon

Update GainsUpdate Gains Update GainsUpdate Gains

Coincidence Timing CalibrationCoincidence Timing Calibration

Purpose:Purpose:•• Adjust for timing delays so events from all blocks are Adjust for timing delays so events from all blocks are

timestampedtimestamped equivalentlyequivalently•• Acquired weekly or after detector maintenanceAcquired weekly or after detector maintenance

γγγγ

γγγγ

Coincidence ?Register

EventYes

Coincidence Timing CalibrationCoincidence Timing Calibration

Blank scan created with incorrect CTC calibrationBlank scan created with incorrect CTC calibration

Daily Quality AssuranceDaily Quality Assurance

Can the Scanner be Used Today?Can the Scanner be Used Today?

Potential ProblemsPotential Problems•• System stability, driftsSystem stability, drifts•• Detector module / PMT/ preamp failureDetector module / PMT/ preamp failure•• Loose cables, connectorsLoose cables, connectors•• Inoperable gantry motors, source loaderInoperable gantry motors, source loader

Daily QA RegimenDaily QA Regimen•• Automatic PMT gain adjustmentAutomatic PMT gain adjustment•• Blank scanBlank scan

•• QA will detect but not prevent these problemsQA will detect but not prevent these problems

4

Example: NMSE and Block BiasExample: NMSE and Block Bias

0.000

0.002

0.004

0.006

0.008

0.010

10-Dec-02 29-Jan-03 20-Mar-03 9-May-03 28-Jun-03 17-Aug-03

Date

Nor

mal

ized

Mea

n S

quar

e E

rror

-8

-6

-4

-2

0

2

4

6

Blo

ck B

ias

(%)

Global NMSE

Slice NMSEBlock Bias

QA must be sensitive to modes of failure of the scannerQA must be sensitive to modes of failure of the scanner

NMSE: Error with respect to baseline scanNMSE: Error with respect to baseline scan

NEMA NU 2-2001NEMA NU 2-2001

PET Performance MeasurementsPET Performance Measurements

National Electrical Manufacturers Association. National Electrical Manufacturers Association. NEMA Standards Publication NU 2NEMA Standards Publication NU 2--2001: 2001: Performance Measurements of Positron Performance Measurements of Positron

Emission Emission TomographsTomographs

NEMA NU 2-2001NEMA NU 2-2001

Acceptance TestingAcceptance TestingCan we use our new scanner?Can we use our new scanner?

Annual QAAnnual QAIs the scanner still performing within specification?Is the scanner still performing within specification?

NEMA NU 2-2001NEMA NU 2-2001

•• Measures performance of PET scanners under conditions that Measures performance of PET scanners under conditions that attempt to represent whole body studiesattempt to represent whole body studies

–– Phantom of length greater than axial FOVPhantom of length greater than axial FOV

–– Out of field activityOut of field activity

•• Standardizes oblique LOR manipulation in 3D acquisitionsStandardizes oblique LOR manipulation in 3D acquisitions

•• DaubeDaube--Witherspoon ME, et al. PET Performance Measurements Witherspoon ME, et al. PET Performance Measurements Using the NEMA NU 2Using the NEMA NU 2--2001 Standard. 2001 Standard. J J NuclNucl MedMed 2002:13982002:1398--1409.1409.

5

Phantoms for NU 2-2001Phantoms for NU 2-2001

NEMA NU 2NEMA NU 2--20012001Three phantoms:Three phantoms:

Scatter phantom Scatter phantom 203x700mm phantom with activity203x700mm phantom with activity

in line sourcein line source

Image quality phantomImage quality phantom

Sensitivity phantomSensitivity phantom

NEMA NU 2-2001NEMA NU 2-2001

Performance Measures:Performance Measures:

•• Spatial Resolution (Transaxial, axial)Spatial Resolution (Transaxial, axial)

•• SensitivitySensitivity

•• Scatter FractionScatter Fraction

•• Count LossesCount Losses

•• Count Rate Correction AccuracyCount Rate Correction Accuracy

•• Image QualityImage Quality

Spatial ResolutionSpatial Resolution

•• Spatial resolution of a system represents its ability to distingSpatial resolution of a system represents its ability to distinguish uish between two points of radioactivitybetween two points of radioactivity

•• FF--18 point sources in air at six locations:18 point sources in air at six locations:

–– (0,1), (0,10) and (10,0) cm (0,1), (0,10) and (10,0) cm

–– Center of axial FOV and Center of axial FOV and ¼¼ axial FOV from centeraxial FOV from center

•• Reconstruct: image pixel < Reconstruct: image pixel < ⅓⅓ expected FWHM expected FWHM

•• Profile width ~ 2 times FWHMProfile width ~ 2 times FWHM

•• Report FWHM and FWTM in radial, tangential and axial Report FWHM and FWTM in radial, tangential and axial directionsdirections

Spatial ResolutionSpatial Resolution

Transaxial Resolution Transaxial Resolution

FWHM @ 1cmFWHM @ 1cm FWHM @ 10cmFWHM @ 10cm

Philips AllegroPhilips Allegro 5.5 mm5.5 mm 5.6 mm5.6 mm

GE DSTGE DST 6.26.2 6.76.7

CTI CTI HiREZHiREZ 6.56.5 7.57.54.64.6 5.85.8

6

SensitivitySensitivity

•• Sensitivity of a scanner represents its ability to detect Sensitivity of a scanner represents its ability to detect annihilation radiationannihilation radiation

•• Rate of true coincidence counts per unit radioactivity (expresseRate of true coincidence counts per unit radioactivity (expressed d in cps/in cps/kBqkBq) in absence of attenuating media) in absence of attenuating media

•• Rationale: need material around source to ensure annihilation ofRationale: need material around source to ensure annihilation ofpositrons, but this material also attenuates the annihilation positrons, but this material also attenuates the annihilation photonsphotons

•• Based on technique by Bailey DL, Jones T, et al. Based on technique by Bailey DL, Jones T, et al. EurEur J J NuclNuclMedMed 1991;18:3741991;18:374--379.379.

SensitivitySensitivity

•• Successive measurements with a 700 mm line source with a known Successive measurements with a 700 mm line source with a known amount of Famount of F--18 surrounded by nested, known absorbers18 surrounded by nested, known absorbers

•• The count rate with no absorber is extrapolated from these measuThe count rate with no absorber is extrapolated from these measuresres

0.0E+00

5.0E+04

1.0E+05

1.5E+05

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4

Sleeve Thickness (cm)

Cou

nt R

ate,

R (

cps)

R0

Activity

RySensitivitSystem 0=

SensitivitySensitivity

0.00

0.01

0.02

0.03

0.04

0.05

0 10 20 30 40 50

Slice

Sen

sitiv

ity (

cps/

kBq)

R0R10

•• Measure at radial locations of 0 and 10cmMeasure at radial locations of 0 and 10cm

•• Report system sensitivity and slice sensitivity profileReport system sensitivity and slice sensitivity profile

0.00

0.10

0.20

0.30

0.40

0 10 20 30 40 50

Slice

Sen

sitiv

ity (

cps/

kBq) R0

R10

2D 3D

SensitivitySensitivity

SensitivitySensitivity(cps/(cps/kBqkBq))

GE DST 2DGE DST 2D 2.02.03D3D 9.39.3

Philips Allegro Philips Allegro 4.44.4

CTI CTI HiRezHiRez 4.54.5

7

Scatter FractionScatter Fraction

•• Scatter fraction is a measure of the system sensitivity to scattScatter fraction is a measure of the system sensitivity to scatterer

TruesScatterScatter

SF+

=

•• Use 203 mm diameter polyethylene cylinder of length 700 mm, withUse 203 mm diameter polyethylene cylinder of length 700 mm, withactivity located in a line source of diameter 2.3mm that is 4.5mactivity located in a line source of diameter 2.3mm that is 4.5mm m off axisoff axis

•• Measured with low activity (Measured with low activity (Randoms:TruesRandoms:Trues < 1%) to avoid random < 1%) to avoid random coincidences, coincidences, deadtimedeadtime and pulse pileup.and pulse pileup. 0

5

10

15

20

25

30

35

40

0 10 20 30 40 50

Slice

Sca

tter

Fra

ctio

n (%

)

Scatter Fraction Scatter Fraction

0

5

10

15

20

0 10 20 30 40 50

Slice

Sca

tter

Fra

ctio

n (%

)

2D 3D

Scatter Fraction Scatter Fraction vsvs SliceSlice

Scatter FractionScatter Fraction

Scatter FractionScatter Fraction LLDLLD((keVkeV))

GE DST 2DGE DST 2D 19%19% 3753753D3D 44%44% 375375

Philips Allegro Philips Allegro 40%40% 410410

CTI CTI HiRezHiRez 36%36% 425425

Count Rate PerformanceCount Rate Performance

•• Measurement of count rate performance gives an indication of Measurement of count rate performance gives an indication of scanner performance as a function of activityscanner performance as a function of activity

•• Use 700 mm long polyethylene cylinder Use 700 mm long polyethylene cylinder

•• Measured with high initial activity of FMeasured with high initial activity of F--1818

3D: 800 3D: 800 MBqMBq; 2D: 5 ; 2D: 5 GBqGBq

•• Acquire data until Acquire data until randomsrandoms and and deadtimedeadtime losses are negligible losses are negligible (14 to 18 hrs)(14 to 18 hrs)

8

Count Rate PerformanceCount Rate Performance

•• Calculate Noise Equivalent Count RateCalculate Noise Equivalent Count Rate

RandomsScatterTrues

TruesNEC RkRR

RR

⋅++=

2

•• Report peak NEC and effective activity concentration at peakReport peak NEC and effective activity concentration at peak

RRNECNEC: Figure of merit relating scanner performance to sinogram : Figure of merit relating scanner performance to sinogram SNR after SNR after randomsrandoms and scatter corrections.and scatter corrections.

For NEMA, k = 1 (calculated For NEMA, k = 1 (calculated RandomsRandoms); R); RNEC NEC (1R)(1R)

3D Count Rate Performance3D Count Rate Performance

0

500

1000

1500

2000

2500

3000

3500

0 10 20 30 40 50 60

Activity Concentration (kBq/cc)

Co

unt R

ate

(kcp

s)

PromptsTrues

R+S

Peak Peak RRTruesTrues: 515 : 515 kcpskcps @ 44 @ 44 kBqkBq/ml/ml

Peak RPeak RNECNEC (1R):117 (1R):117 kcpskcps @ 20 @ 20 kBqkBq/ml/ml

0

20

40

60

80

100

120

140

0 10 20 30 40 50 60

Activity Concentration (kBq/cc)

NE

CR

(kc

ps)

NECR (1R)NECR (2R)

Count Rate PerformanceCount Rate Performance

Count Rate PerformanceCount Rate Performance

Peak RPeak RNECNEC ActivityActivity((kcpskcps)) ((kBqkBq/ml)/ml)

CTI CTI HiREZHiREZ 8585 2121

GE DST 2DGE DST 2D 8484 49493D3D 6363 1212

Philips AllegroPhilips Allegro 3030 99

Image Quality MeasurementImage Quality Measurement

•• Standardized imaging situation that simulates a clinical whole Standardized imaging situation that simulates a clinical whole body imaging conditionbody imaging condition

•• Phantom consists of a torso phantom with hot and cold lesions inPhantom consists of a torso phantom with hot and cold lesions ina warm backgrounda warm background

•• Scatter phantom abutted to image quality phantomScatter phantom abutted to image quality phantom

9

Image Quality MeasurementImage Quality Measurement

•• Hot spheres: 10, 13, 17, 22 mm idHot spheres: 10, 13, 17, 22 mm id

•• Cold spheres: 28, 37 mm idCold spheres: 28, 37 mm id

•• Lung insertLung insert

•• Activity in hot spheres 8 and 4 times that of backgroundActivity in hot spheres 8 and 4 times that of background

•• Activity in background 5.3 Activity in background 5.3 kBqkBq/ml/ml

•• Simulated acquisition 100cm in 60minSimulated acquisition 100cm in 60min

•• Repeat acquisition three timesRepeat acquisition three times

•• Reconstruct using clinical protocolReconstruct using clinical protocol

min7cm100

min60 ≈∆= zTacq

Image Quality MeasurementImage Quality Measurement

•• Report image contrast and Report image contrast and SNRsSNRs for hot and cold lesions, for hot and cold lesions, residual error in lung, variability in backgroundresidual error in lung, variability in background

•• Visual inspection for artifactsVisual inspection for artifacts

NEMA and Lutetium-176NEMA and Lutetium-176

•• LSO is inherently radioactiveLSO is inherently radioactive

•• Background radiation gives rise to Background radiation gives rise to RandomsRandoms, some , some TruesTrues

•• Implications for NEMA: cannot obtain Implications for NEMA: cannot obtain Randoms:TruesRandoms:Trues ratio of < 1%ratio of < 1%

•• For count rate and sensitivity measurements For count rate and sensitivity measurements –– acquire background acquire background scan to estimate intrinsic true and random ratesscan to estimate intrinsic true and random rates

•• Watson CC, et al. Watson CC, et al. J J NuclNucl MedMed 2004;45:8222004;45:822--826.826.

•• Erdi YE, et al. Erdi YE, et al. J J NuclNucl MedMed 2004;45:8132004;45:813--821.821.

SummarySummary

•• Need for QA regimenNeed for QA regimen

•• Quantitative analysis of QA dataQuantitative analysis of QA data

•• NEMA NU 2NEMA NU 2--2001 Performance Standard procedures2001 Performance Standard procedures