Automatic three-dimensional multimodality registration using radionuclide transmission CT attenuation maps: A phantom study

Coregistration of images from a single subject, acquired by different modal ities, is important in clinical diagnosis, surgery and therapy planning. Th e purpose of this study was to evaluate, using a physical torso phantom, a novel, fully automated method for three-dimensional image registration of C T and SPECT, using radionuclide transmission (RNT) attenuation maps. Method s: We obtained CT scans and SPECT scans paired with RNT maps of an anthropo morphic cardiac phantom, RNT attenuation maps were acquired using an uncoll imated Tc-99m-filled flood source. RNT and SPECT scans were acquired in the same spatial orientation (usual clinical practice in nonuniform attenuatio n correction). In addition, CT attenuation maps (CTMAPs) for Tc-99m SPECT w ere generated from CT by linear energy scaling. RNT maps were registered to CT and CTMAPs by iterative simplex minimization of count difference and un iformity index (sum of RNT map intensity variances corresponding to each in tensity level in the CT volume). In each iteration, three shifts and three angles were adjusted. To register SPECT to CT, we applied the RNT transform ation parameters to SPECT. Results: RNT maps could be registered to CT and CTMAP images using both criteria. The average three-dimensional distance be tween landmark and automated registration was 2.5 +/- 1.2 mm for count diff erence and 3.3 +/- 1.3 mm for uniformity index. The three-dimensional repro ducibility errors were 1.2 +/- 0.7 mm for count difference, 2.1 +/- 0.5 mm for uniformity index and 2.3 +/- 1.0 mm for manual marker registration. The minimization of uniformity index was robust when up to 50% CT or RNT slice s were missing and was not affected significantly (<2 mm) by realistic vari ation in CT values (+/-12 Hounsfield units). Conclusion: In addition to typ ical use in nonuniform attenuation correction, RNT maps can be used for ful ly automated three-dimensional registration of SPECT to CT. Such registrati on is not affected by features and quality of SPECT images and avoids diffi culties associated with fiducial markers. Our method can be applied to SPEC T-CT registration of various organs, such as brain, heart, lungs, breasts a nd abdomen, including oncological scans.