Evaluation of the effect of reconstructed image pixel size on defect detectability in Tl-201 fan-beam SPECT by an observer performance study

The effect of reconstructed image pixel size in the fan-beam filtered backprojection method in myocardial defect detection was investigated using an observer performance study and receiver operating characteristics (ROC) analysis. A mathematical phantom of the human torso was used to model the anatomy and Thallium-201 (Tl-201) uptake in humans. Realistic projections from the phantom were simulated using a low-energy high resolution fan-beam collimator that incorporated the effects of photon attenuation, spatially varying detector response, scatter, and Poison noise. For a fan-beam collimator with a focal length of 55 cm and with a radius of rotation of 25 cm, the magnification at the center of rotation was two and the maximum magnification in the reconstructed region of interest was three. Myocardial defects were simulated as Gaussian-shaped decreases in Tl-201 uptake distribution. By changing the reconstructed image pixel size, five different classes of reconstructed images resulted, with projection bin width to reconstructed image pixel size (PBIP) ratios of 1, 2, 3, 4, and 5. The results from the observer study indicate that the reconstructed image pixel size has a significant effect on myocardial defect detection in reconstructed Tl-201 SPECT images. Moreover, the study indicated that in order to ensure maximum defect detectability the PBIP ratio should be at least as large as the maximum possible magnification within the reconstructed image array.<
>