We describe and validate a volumetric three-dimensional registration method, and compare it to our previously validated two-dimensional/three-dimensional method. CT/MRI and SPECT data from 14 patients were interactively fused using a polynomial warping technique. Registration accuracy was confirmed visually and by a nonsignificant F value from multivariate analysis of the transformed landmarks, a significant difference of the squared sum of intensity differences between the transformed/untransformed and the reference volume both at the 0.05 (p > 0.05) confidence level and an average 31% improvement of the correlation coefficient and cross correlation. For the two-dimensional/three-dimensional method, ROI center-to-center distance ranged from 1.42 to 11.32 mm (for liver) with an average of 6.13 mm +/- 3.09 mm. The average ROI overlap was 92.51% with a 95% confidence interval of 90.20-96.88%. The new method is superior because it operates on the true three-dimensional volume. Both methods give good registration results, take 10 to 30 min, and require anatomic knowledge.

Although the use of F-18 FDG CD shows great promise for the identification of tumors, it shares the same drawbacks as those associated with radiolabeled monoclonal antibody SPECT and ligand-based positron emission tomographic scans in that anatomic markers are limited. This study shows that image registration is feasible and may improve the clinical relevance of CD images.

Combined PET/CT scanners provide the ability to produce matching metabolic (from PET) and anatomic (from CT) information in a single examination. However, misalignments continue to exist in tumor localization in PET and CT images acquired using these scanners, due to their inability to compensate for nonrigid misalignment resulting from patient breathing and involuntary movement. We demonstrate that our automatic image subdivision-based elastic registration algorithm can correct this misalignment. In a quantitative validation involving 13 expert-identified tumor nodules in six PET-CT image pairs, the algorithm demonstrated statistically significant improvement over the scanner-defined localization. The accuracy of algorithm-determined localization was evaluated to be comparable to average manually defined localization. The results indicate the potential of using our registration algorithm for applications like radiotherapy treatment planning and treatment-monitoring involving combined PET/CT scanners.

No abstract

No abstract