학술논문

SPECT dual-isotope myocardial perfusion imaging with a 20-pinhole collimator: A simulation study
Document Type
Conference
Source
IEEE Nuclear Science Symposuim & Medical Imaging Conference Nuclear Science Symposium Conference Record (NSS/MIC), 2010 IEEE. :3029-3031 Oct, 2010
Subject
Nuclear Engineering
Engineered Materials, Dielectrics and Plasmas
Bioengineering
Power, Energy and Industry Applications
Components, Circuits, Devices and Systems
Computing and Processing
Communication, Networking and Broadcast Technologies
Collimators
Stress
Sensitivity
Cameras
Myocardium
Monte Carlo methods
Single photon emission computed tomography (SPECT)
multipinhole
myocardial perfusion imaging (MPI)
dual-isotope imaging
Language
ISSN
1082-3654
Abstract
Single photon emission computed tomography (SPECT) has essentially remained an unchanged technology since the introduction of parallel-hole collimation in 1964 and stands to gain significantly from sensitivity improvements. Benefits from improved sensitivity include shorter acquisition times and smaller dose requirements. An order of magnitude increase in sensitivity over conventional technology is possible with the use of multipinhole collimators on conventional SPECT cameras. Here we present MGEANT Monte Carlo simulation results of a dualisotope myocardial perfusion imaging study performed with a pair of focusing 20-pinhole collimators with tungsten apertures demonstrating that a complete rest/stress study is possible in 10 minutes with a reduced 99m Tc dose. Two radionuclides were simulated, 99m Tc (500 μCi, rest) and 201 Tl (200 μCi, stress), with 20% and 30% window sizes, respectively. Full isotropic emission was simulated for 8 gamma-ray lines. The imaging protocol consisted of 2 views per head with 300 s per view. The activity distributions were derived from the mathematical cardiac and torso (MCAT) phantom. Attenuation was modeled using a water phantom. Images were reconstructed using 75 iterations of pixel-based ordered subsets expectation maximization and post-filtered with an 12.5 mm (FWHM) 3D Gaussian filter. Scatter corrections were performed using the triple energy window method. Scatter comprises 37% and 61% of the total counts in the 99m Tc and 201 Tl energy windows, respectively. 99m Tc downscatter contributes 52% of the scatter component in the 201 Tl window, though tungsten fluorescence is negligible (0.2%). Reconstructed images are qualitatively in agreement with the input activity distributions.