Increase of accuracy in intraoperative navigation through high-resolution flat-panel volume computed tomography: experimental comparison with multislice computed tomography-based navigation.

HYPOTHESIS: High-resolution imaging, as provided by flat-panel-based volume computed tomography (fpVCT), could increase navigation accuracy and could therefore improve image-guided procedures or make novel navigated surgery concepts possible. BACKGROUND: Intraoperative navigation is an accepted tool in head and neck surgery. However, its use is limited in the lateral cranial base because of its low surgical accuracy. Surgical accuracy is substantially influenced by the resolution of the underlying data set. The fpVCT offers a resolution of nearly two times higher than multislice computed tomography (MSCT). Target registration error (TRE), as a measurement for surgical navigation accuracy, should decrease when navigation is based on fpVCT data sets. METHODS: An acrylic glass phantom with 37 fiducial points was scanned in a current MSCT and in an experimental fpVCT. Both data sets were imported in an optical navigation system. Five fiducial points were used for registration, and seven points were used for measuring TRE. The distance between the indicated pointer tip and the corresponding fiducial point in data set was measured as TRE. Registration and TRE measurement were repeated five times for each computed tomographic data set. Average TREs were calculated, and results were compared using t-test. RESULTS: The average TRE using MSCT (0.82 mm [standard deviation, 0.35 mm]) was significantly higher than that using fpVCT (0.46 mm [standard deviation, 0.22 mm]) (p < 0.01). CONCLUSION: Submillimeter surgical navigation accuracy is possible using high-resolution fpVCT. This could be highly beneficial in cranial base surgery navigation.