Receiver operating characteristic evaluation of iterative reconstruction with attenuation correction in Tc-99m-sestamibi myocardial SPECT images

The purpose of this study was to evaluate differences in myocardial defect detection between Tc-99m-sestamibi myocardial SPECT images reconstructed us ing conventional filtered backprojection (FBP) without attenuation correcti on (AC) and those reconstructed using maximum-likelihood expectation maximi zation with nonuniform attenuation correction (MLAC). Methods: An observer study and receiver operating characteristic (ROC) curve analysis were perfo rmed using simulated Tc-99m-sestamibi SPECT data from a population of 24 ma thematic anthropomorphic torso phantoms, which realistically modeled a wide range of anatomic variations. The phantoms modeled male patients with a fl at diaphragm, male patients with a diaphragm raised to the level of the hea rt, and female patients with large breasts, Transmural, cold defects with a contrast of 0.25 were simulated in the left ventricular wall for 6 locatio ns. Noisy projection data were generated from the phantoms and included the effects of nonuniform attenuation, collimator-detector response, and scatt er. The data were then reconstructed using FBP and MLAC. Images were displa yed in the short- and long-axis formats, as in clinical practice. Eight obs ervers viewed blocks of FBP and MLAC images and, for each image, indicated on a continuous rating scale the probability that a defect was present. Fro m the rating data, FBP and MLAC ROC curves were generated, and their areas (A,) were estimated and compared. Results: In general, the FBP and MLAC ROC curves did not cross and the MLAC curve showed a higher A, than did the co rresponding FBP curve. For male phantoms with a flat diaphragm, the average difference in A,was 0.04 and was not statistically significant (at the P = 0.05 level) for 6 of 8 observers. For male phantoms with a raised diaphrag m, the average difference in 8, was 0.22 and was statistically significant for 6 of 8 observers. For female phantoms with large breasts, the average d ifference in A, was 0.19 and was statistically significant for all 8 observ ers. Conclusion: This study showed an improvement in defect detection in my ocardial SPECT images using MLAC in comparison with images using FBP withou t AC, particularly for patients with large breasts or with a diaphragm rais ed to the level of the heart.