Quantitative imaging of cardiovascular function with pinhole SPECT in mice and rats – Effects of attenuation and scatter

Abstract

1769

Objectives: Both microSPECT and clinical SPECT systems have been used for molecular imaging of small animals. Attenuation is not corrected with clinical systems as they lack transmission imaging with sufficient spatial resolution to produce detailed attenuation maps for small animals; moreover, it is generally believed that the animal’s small size makes photon attenuation negligible for energies at 140 and 159 keV. We performed a study to determine the errors due to photon attenuation and scatter for myocardial imaging of mice and rats using pinhole collimators with clinical SPECT systems. Methods: Computer simulations were performed using the 4D digital MOBY mouse phantom which was scaled appropriately to simulate the size of a rat. MOBY provides a realistic emission distribution and attenuation map from which 180 projections over 360° were simulated. A 128×128 reconstruction without attenuation correction was compared to a reconstruction with attenuation correction. In addition, a SimSET Monte Carlo simulation was performed to obtain the scatter fraction for water filled cylindrical phantoms the size of a mouse and a rat. Results: At 140 keV, the reconstruction of a mouse myocardium had a mean value that was 11% less without attenuation correction than that obtained with attenuation correction and for the rat the mean reconstructed value was 25% less. The underestimation of radionuclide uptake is fairly uniform over the myocardium and does not introduce dramatic regional variations. For 140 keV the scatter fraction is 10% for a cylinder the size of a rat and 7% the size of a mouse. The scatter fractions likely will be smaller in animals due to reduced attenuation in lung tissue which is not modeled in the uniform water-filled cylindrical phantom. Conclusions: Quantitatively accurate reconstructions of myocardial images require attenuation correction for imaging rats, but are not as critical for mice. Scatter does not contribute significant errors in myocardial pinhole SPECT imaging in small animals at photon energies of 140 keV.

Research Support (if any): NIH R01-EB00121, R01-HL50663, and R01-EB00348; and DOE Contract DE-AC02-05CH11231.