Performance characteristics of a small animal PET camera for - PowerPoint PPT Presentation

Performance characteristics of a small animal PET camera for molecular imaging Hastings DL 1 , Reader AJ 2 , Julyan PJ 1 , Zweit J 2,3 Jeavons AP 4 , Jones T 5 1 North Western Medical Physics, Christie Hospital, Manchester 2 Chemical Engineering and Analytical Science, University of Manchester 3 Paterson Institute for Cancer Research, Manchester 4 Oxford Positron Systems, Oxfordshire 5 Manchester Molecular Imaging Centre, University of Manchester.

HIDAC Detector Module 511 keV

Quad HIDAC Detector

Quad HIDAC Camera 13 million sub-millimetre detectors

Quad HIDAC Camera 4 million sub-millimetre detectors

Spatial Resolution Line source glass/polyamide tube – 50µm internal diameter – 300µm external diameter 22 Na concentration of 75 kBq/ml – Iterative reconstruction – 0.25mm voxels – 16 iterations – 1 subset – no resolution modeling FWHM determined according to NEMA

Spatial Resolution Transaxial 1.1 1 0.9 FWHM (mm) 0.8 0.7 Horizontal Vertical 0.6 Horizontal Mean = 1.02 Vertical Mean = 1.00 0.5 -100 -50 0 50 100 Distance along Axis (mm)

Spatial Resolution Axial Resolution 1.1 1 0.9 FWHM (mm) 0.8 0.7 Axial: Mean = 0.99mm 0.6 0.5 -80 -40 0 40 80 Radial Distance from Centre (mm)

Spatial Resolution FWHM (mm) Transaxial Transaxial Axial Horizontal Vertical Mean 1.02 1.00 0.99 Max 1.09 1.05 1.05 Min 0.93 0.86 0.88 Standard 0.04 0.04 0.05 Deviation

Sensitivity • Point source of 18 F - 175.2 kBq • Electroplated 18 F on tip of an Al wire • 5mm blu-tack ball on tip as positron annihilator • Iterative reconstruction – 1mm voxels – length = 250mm; diameter = 150mm – regularization = 0 – resolution recovery = none – Iterations = 1, subsets = 32

Sensitivity Point Source and Profile PET image 8.0E+05 6.0E+05 Counts 4.0E+05 2.0E+05 0.0E+00 40 60 80 100 Position (mm)

Sensitivity Events acquired S = × 100 % = 0.95% Number of positron emissions during scan time S c = sensitivity corrected for scatter background = 0.75% • Bin data into 2D parallel projections (bin size 0.5mm) • Integrate over the 2D plane in the axial ( z ´) profile p i 100000 single 1D profile p i (for i =1… I bins) – 8.0E+05 • Trues are in +/- 1.5mm of peak 10000 6.0E+05 • Ratio of trues to scatter-background 1000 Counts 4.0E+05 ∑ 100 p i 2.0E+05 i ∈ T = 0.786 10 I ∑ p 0.0E+00 i 40 60 80 100 1 Position (mm) i = 1 40.0 60.0 80.0 100.0 120.0 140.0 Position (mm)

Count Rate Capability • Cylindrical phantom to simulate a mouse – diameter 30mm, length 50mm – start activity = 74 MBq 18F • List mode acquisition for 21 hours – final activity = 0.04MBq – refresh rate 120 seconds – total coincidences, singles, randoms rates recorded

Count Rate Capability Singles 1400 Count Rate - Singles 1200 Detector 1 1000 Detector 2 Detector 3 800 Detector 4 kcps 600 400 200 0 0 10 20 30 40 50 60 70 Activity (MBq)

Count Rate Capability Noise Equivalent Count Rate ( ) ( ) T Totals R SF = − × 1 − SF = scatter fraction = 29% (Missimer 2004 * ) T NECR = T = trues rate 2 kR S 1 + + = scatter rate S T T R = randoms rate ( ) S SF Totals R = × − k = proportion of the object subtended by the imaging device (?? 0.10) * Missimer et al, Phys Med Biol. 49 (2004) 2069-2081

Count Rate Capability Coincidences 400 Count Rate - Coincidences Trues Total 20% loss at Randoms 300 Trues 11.5 MBq Scatter (SF=29%) NEC (ko=0.10) kcps 200 100 0 0 10 20 30 40 50 60 70 Activity (MBq)

Count Rate Capability 60.0 Count Rate - Coincidences Totals Total 50.0 Randoms Trues 40.0 Trues kcps 30.0 20.0 Randoms 10.0 0.0 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0 Activity (MBq)

Energy Dependence of Efficiency 4 511 (% Counts/emitted gamma) 3 F18 Singles Sensitivity 1253 364 Co60 I131 784 2 Na22 Module 1 Module 2 1 Module 3 141 Module 4 Tc99m 0 0 200 400 600 800 1000 1200 1400 Energy (keV)

Summary of Results Parameter Measured Spatial Resolution Transaxial horiz 1.02 (0.04) Transaxial vert 1.00 (0.04) Axial 0.99 (0.05) Senstivity Uncorrected = 0.95% Corrected = 0.75% Count Rate 20% loss rate at 11.5MBq Spatial Calibration Deviation = 0.0 mm

PET Images of 18-FDG Mouse - 1 Transaxial Coronal Sagittal Max Intensity Projection Mouse injected with 18-FDG Image reconstructed into 0.5mm voxels Slice thickness 0.5mm Transaxial, Coronal & Sagittal slices centred on the heart

18F-Fluoride Bone Images in Mouse Sagittal Coronal Courtesy: Munster University Hospital

It’s a long road ….

It’s a long road ….