QUANTITATIVE VALIDATION OF A NEW COREGISTRATION ALGORITHM

A new coregistration software package, Neuro900 Image Coregistration s oftware (Strichman Medical Equipment), has been developed specifically for nuclear medicine. With this algorithm, the correlation coefficien t is maximized between volumes generated from sets of transaxial slice s. No localization markers or segmented surfaces are needed. The coreg istration program was evaluated for translational and rotational regis tration accuracy. A Tc-99m HM-PAO split-dose study (0.53 mCi low dose, L, and 1.01 mCi high dose, Il) was simulated with a Hoffman Brain Pha ntom with five fiducial markers. Translation error was determined by a shift in image centroid, and rotation error was determined by a simpl ified two-axis approach. Changes in registration accuracy were measure d with respect to: 1) slice spacing, using the four different combinat ions LL, LH, HL, HH, 2) translational and rotational misalignment befo re coregistration, 3) changes in the step size of the iterative parame ters. In all the cases the algorithm converged with only small differe nce in translation offset, theta and phi. At 6 mm dice spacing, transl ational errors ranged from 0.9 to 2.8 mm (system resolution at 100 mm, 6.8 mm). The converged parameters showed little sensitivity to count density. In addition the correlation coefficient increased with decrea sing iterative step size, as expected. From these experiments, we foun d that this algorithm based on the maximization of the; correlation co efficient between studies was an accurate way to coregister SPECT brai n images.