A novel easy-to-use phantom for the determination of MTF in SPECT scanners.

PURPOSE To evaluate modulation transfer function (MTF) in single photon emission computed tomography (SPECT) systems using the line spread function (LSF) method and a novel flood source which can be easily fabricated with materials accessible in hospital facilities. METHODS A Tc-99m-based flood source (E(γ) = 140 keV) consisting of a radiopharmaceutical bound to the grains of a radiographic film was prepared in laboratory. Various films and radiopharmaceuticals were examined, in order to obtain a thin homogenous and reproducible flood source. The source showing best uniformity and reproducibility was placed between two PMMA blocks and images were obtained by using the brain tomographic acquisition protocol (brain) and the myocardial perfusion tomographic acquisition protocol (heart). MTF was evaluated by determining the LSF for various reconstruction methods and filters. MTF calculation was obtained by the utilization of a custom made software in which a method similar to the one proposed by Boone [Med. Phys. 28, 356-360 (2001)] was implemented. All imaging experiments were performed in a Siemens e-Cam γ-camera. Furthermore, MTF was assessed through the point spread function (PSF) following conventional methods. RESULTS The optimum homogeneity was obtained by immersing an Agfa MammoRay HDR Medical x-ray film in a solution of dithiothreitol (DTT, 10(-3) M)/Tc-99m(III)-DMSA (DMSA: trivalent technetium-99m-dimercapto-succinic acid, 40 mCi/40 ml) for 30 min in the dark. These films exhibited better uniformity (CV < 1.9%). Higher MTF values were obtained for the brain scan protocol with iterative 3D with eight iterations reconstruction method. MTF of the brain protocol was in all cases better than the heart protocol. MTFs derived from LSF were more precise compared with those obtained from PSF since their reproducibility was better in all cases, providing a mean standard deviation of 0.0065, in contrary to the PSF method which gave 0.0348. CONCLUSIONS The method presented here is novel and easy to implement, requiring materials commonly found in clinical practice. Furthermore, this technique which is based on the LSF method reduces measurement noise levels due to the larger amount of data averaging than in the conventional PSF method. Furthermore, MTF can be assessed easily, in three dimensions (3D), by placing the flood source either in sagittal or coronal direction.