Positron Emission Tomography - PET

Positron Emission Tomography - PETPositron Emission Tomography - PETPositron Emission TomographyPositron Emission Tomography (PET): Coincidence detection ofannihilation radiation from positron-emitting isotopes followed bytomographic reconstruction of 3-D activity distribution.Some unique features of PET: Use of “electronic collimation” instead of lead collimation. High detection efficiency Uniform resolution Accurate attenuation correction “Absolute” Quantification Use of short-lived biologically active ter18FDG18FDOPAM. Dahlbom M284B Winter 20161

Positron Emission Tomography - PETCoincidence DetectionAmp.PHAAmp.511 keV?511 keV?PHACoinc.CounterM. Dahlbom M284B Winter 20162

Positron Emission Tomography - PETPET GantryECAT HR PET GantryM. Dahlbom M284B Winter 20163

Positron Emission Tomography - PETBernard Bendriem, Ph.D.Siemens Hi-Rez / Biograph 16 Hi-RezVice-President, R&DMarch 19, 2004LSO 13x13 elements/block – 4x4x20mm2 detector elementsCoincidence DetectionM. Dahlbom M284B Winter 20164

Positron Emission Tomography - PETCoincidence DetectionN (N -1)# of possible LORs: N N number of detectors LOR2Spatial ResolutionThe spatial resolution in PET is primarily determined by:Detector sizePhysics of positron decaySystem geometryDetector materialM. Dahlbom M284B Winter 20165

Positron Emission Tomography - PETSpatial ResolutionFor a source placed at the midpoint between two scintillationdetectors with a width wd, the geometric line spread function has atriangular shape with a FWHM of wd /2.wddwwFWHM dFWHM wd/22Spatial Resolution -TangentialFor sources located between the midpoint and the detectorsurface the LSF will have a trapezoidal shape with width varyingfrom wd /2 (at the center) and wd at the detector surface.M. Dahlbom M284B Winter 20166

Positron Emission Tomography - PETSpatial Resolution - RadialTransaxial Resolution or ECAT EXACT HR8TangentialAxial section7FWHM tangFWHM radFWHM(mm)6Radial543R 021005101520R (cm)Bernard Bendriem, Ph.D.Small Lesion Detection: A Phantom StudyAll spheres containthe same activityconcentration13March 19, 2004172210Profile (10 mm)2837Recovery (%)Vice-President, R&D1002228371750130Standard8 x 8 detector10Sphere diameterHI-REZ13 x 13 detectorRecovery coefficientsM. Dahlbom M284B Winter 20167

Positron Emission Tomography - PETSpatial ResolutionAlthough the most energetic positrons can travel several mm before annihilating,only a few of these are emitted.The average positron energy emitted is approximately 1/3-1/2 of the maximumenergy.The total path length the positrons travel is not along a straight path. Throughinelastic interactions with electrons in the positrons path is deflected. The distancefrom the mother nucleus is therefore much shorter.Positron RangeFrom Levin & Hoffman PMB 44, 1999M. Dahlbom M284B Winter 20168

Positron Emission Tomography - PETPositron Range 50% 65%18F124I635 keV1.53 & 2.14 MeVNon-colinearity100 cm Ø 2.5 mm FWHM 0.3 FWHM15 cm Ø 0.3 mm FWHMM. Dahlbom M284B Winter 20169

Positron Emission Tomography - PETSpatial ResolutionThe measured resolution (intrinsic resolution) of the system is aconvolution of the various resolution components.If the different resolution components are assumed to beGaussian in shape and are described by a FWHM then thecombined resolution is the squared sum of the individualresolution components:222FWHMtotal FWHMdetector FWHM positron FWHM angulation3D Acquisition PET2D Acquisition PETM. Dahlbom M284B Winter 20163D Acquisition PET10

Positron Emission Tomography - PET3D vs. 2D PETThe main advantage of the 3-D acquisition in PET is an improvedsensitivity of 5-7 times the 2-D sensitivity.The drawback is that the scatter fraction increases by a factor of 3.Non-uniform axial sensitivityHigher Randoms Rates Increased Noise (offsets sensitivity gain)Dead-time problems when using slow detectorsImage reconstruction is more complexMore dataM. Dahlbom M284B Winter 201611

Positron Emission Tomography - PETCoincidence DetectionAmp.PHAAmp.511 keV?511 keV?PHACoinc.CounterTiming Resolution16064 nsCounts12080FWHM 6 ns400020406080100120ChannelTiming spectrum showing the PHA trigger time variation for a pair of BGO detectors incoincidence. The two peaks corresponds to two separate measurements where anadditional delay of 64 ns of the stop pulse for channel-to-time calibration.M. Dahlbom M284B Winter 201612

Positron Emission Tomography - PETCoincidence DetectionAll coincidence detection systems have a finite time resolutionBGONaIGSOLSO 6 ns FWHM 4 ns FWHM 2 ns FWHM 0.5 ns FWHMCoincidence DetectionAmp.PHAAmp.511 keV?511 keV?PHACoinc.CounterM. Dahlbom M284B Winter 201613

Positron Emission Tomography - PETRandom CoincidencesBecause of the finite width of the logic pulses that are fed into thecoincidence circuit, there is a probability for random or accidentalcoincidences between unrelated events.True Single RandomCoinc. Event Coinc.Detector 1TimeDetector 2τTrueCoinc.SingleEventRandom CoincidencesIf N1 and N2 are the individual average count rates of detector 1 and 2,respectively, then it can be shown that the random coincidence rate forthe pair of detectors is:NR 2τ N1 N2Where 2τ is the coincidence window (or τ is the width of the singlespulses)M. Dahlbom M284B Winter 201614

Positron Emission Tomography - PETEvent TypesTrue EventScattered EventRandom EventMultiple EventSignal-to-NoiseTrue Coincidences ActivityGood events!S/N M. Dahlbom M284B Winter 2016TT15

Positron Emission Tomography - PETSignal-to-NoiseRandom Coincidences Activity2Can be accurately corrected forCorrection increases image noiseDetector material dependentS /N TT 2RSignal-to-NoiseScattered Coincidences ActivityReduces Image ContrastRequires correctionAnalytical estimationCorrection increases image noiseS/N M. Dahlbom M284B Winter 2016TT S 2R16

Positron Emission Tomography - PETSignal-to-NoiseMultiple Coincidences: Activity3Never savedSource of Dead timeImprovements in PET Image QualityPET III1975NaIECA T II1976NeuroECAT1978ECA T 9311985BGOECA T EXACT HR 1994CTI/SiemensM. Dahlbom M284B Winter 201617

Positron Emission Tomography - PETPET DetectorsMost modern PET system use a different detectortechnology where a large number of scintillationcrystals are coupled to a smaller number of PMTs.In the block detector, a matrix of cuts are madeinto a solid block of scintillator material to definethe detector elements.The depth of the cuts are adjusted to direct thelight to the PMTs.The light produced in each crystal, will produce aunique combination of signals in the PMTs, whichwill allow the detector to be identified.Bernard Bendriem, Ph.D.The Technology : HiRezStandard Detector6.4 mm x 6.4 mm64 crystals/block144 blocks/scanner9216 crystals/scanner3.4 mm slice width47 slicesM. Dahlbom M284B Winter 2016Vice-President, R&DMarch 19, 2004HI-REZ Detector4.0 mm x 4.0 mm169 crystals/block144 blocks/scanner24336 crystals/scanner2 mm slice width81 slices18

Positron Emission Tomography - PETScintillator MaterialsNaI (Tl)BGOGSOLSOLYSOLaBr3Density [g/ml]3.677.136.717.357.15.291/µ [cm]2.881.051.431.161.2 2Index oNoNoNoYesRuggedNoYesNoYesYesYesPeak Emission[nm]410480430420420380Decay Constant[ns]23030060404125Light Output10015357575 100EnergyResolution7.8208.9 9117.556Presentation Title Goes In This AreaM. Dahlbom M284B Winter 201619

Positron Emission Tomography - PET57Presentation Title Goes In This AreaImprovements in PET Image QualityLSOECAT HRRTCTI/SiemensM. Dahlbom M284B Winter 201620

Positron Emission Tomography - PETCorrections in PETIn most nuclear medicine procedures, the goal is to produce an image inwhich the gray scale or count density is directly proportional to theregional isotope concentration.In order to achieve this in PET it is necessary to apply a number ofcorrections:Attenuation of photons in tissueNon-uniform response of detector elementsRandom coincidence eventsDetection of scattered eventsLoss of counts at high count rates - dead-timeIsotope decayAbsolute Calibration & cross calibration with other instrumentsHow accurate these corrections are will have a direct impact on thequantitative measurement.Attenuation CorrectionM. Dahlbom M284B Winter 201621

Positron Emission Tomography - PETAttenuation CorrectionAttenuation CorrectionIn PET imaging of the brain, the shape of the head can be approximatedwith an ellipse. The dimensions of the fitted ellipse can be estimated byfirst reconstructing the data without attenuation correction. Then anellipse is drawn onto the image from which the attenuation correction canbe derived. The attenuation correction is the applied to the data and theimage is reconstructed again.This method can be fairly time consuming, especially on systemproducing a large number of transaxial slices.Atten. Corr. eµDM. Dahlbom M284B Winter 201622

Positron Emission Tomography - PETAttenuation Correction68GeSourceBlank ScanTransmission ScanWithout Image SegmentationWith Image SegmentationM. Dahlbom M284B Winter 201623

Positron Emission Tomography - PETAEarly framenon-AC EMOriginal TXFused TX-EM(Match)Early frameAC EMBLate framenon-AC EMOriginal TXFused TX-EM(Mismatch)Late frameAC EM(before MC)CLate framenon-AC EMTX after MCFused TX-EM(Match)Late frameAC EM(after MC)M. Dahlbom M284B Winter 201624

Positron Emission Tomography - PETPET/CTGEPhilipsSiemensx 0.495H.U. µ70 keVx 0.406M. Dahlbom M284B Winter 201625

Positron Emission Tomography - PETTime-of-flight PETRDet 2RDet 1xR-xR xs v tR x vt1R x vt22x v(t2 t1 ) x cΔtM. Dahlbom M284B Winter 2016226

Positron Emission Tomography - PETTime-of-flight PETFor ideal detectors, TOF would eliminate the need for imagereconstruction, since the measurement would allow eachevent to be accurately positioned in space.All detectors have a finite time resolution, or uncertainty intiming. This translates to an uncertainty in positioning.BGO 5 nsNaI 1.5 nsCsF, LaBr3 0.45 nsBaF2, LSO, LYSO 0.3 ns75 cm22.5 cm6.7 cm4.5 cmTime-of-flight PETFigure 1. Image elements contributing to a LOR, forconventional PET (left) and TOF PET (right).M. Dahlbom M284B Winter 201627

Positron Emission Tomography - PETTime-of-flight PETEven with a finite time resolution, using the TOF information animprovement in signal-to-noise ratio (S/N) can be achieved:SNRTOF DSNRconv. Δx2DSNRconv.cΔtTime-of-Flight vs. Conventional PETBetter information sent to reconstructionConventional PETImage FormationTruthTime-of-FlightImage FormationMore precise localization of annihilation event improves image qualityM. Dahlbom M284B Winter 201628

Positron Emission Tomography - PETTime-of-flight PET - 1980’sProblems with TOF in the 80’sPoor detection efficiency of available scintillatorsTOF Gain did not offset the poor efficiencyTo improve the efficiency, large detector modules were usedA more significant gain in S/N could be achieved by using highresolution detectors and conventional detection methods (Phelps,Hoffman, Huang, 1982).Time-of-flight PET - 2006Scintillators:CsF, BaF2 LSO, LYSO - fast, high light, and denseDetectors/PMTs:1:1 coupling 100:1 crystal encoding - spatial resolutionGeometry:2D (septa) 3D with large axial FOV - sensitivityReconstruction:Analytic (FBP) iterative (list-mode) - system modelingElectronics:Accurate and stableM. Dahlbom M284B Winter 201629

Positron Emission Tomography - PETCan we see TOF improvement?non TOFTOF5 min3 min1 min6-to-1 contrast; 35-cm phantomJ. Karp, U of PennNoticeable improvement with TOF with large size phantomGemini TF - patient studyRectal carcinoma, metastases inmesentery and bilateral iliac chainsnon-TOF114 kg; BMI 38.112 mCi; 2 hr post-inj3min/bedTOFJ. Karp, U of PennLesion contrast (SUV) improves with TOF reconstructionM. Dahlbom M284B Winter 201630

Positron Emission Tomography - PETME Phelps et. al.DH Silverman et. al.M. Dahlbom M284B Winter 201631

Positron Emission Tomography - PETDH Silverman et. al.DH Silverman et. al.M. Dahlbom M284B Winter 201632