The effect of intravenous contrast on accuracy of Cyberknife skull tracking

Objective: To quantitatively evaluate the effect of using digitally reconstructed radiograph (DRR) generated images from contrast-enhanced CT on the accuracy of Cyberknife skull tracking. Methods: Forty-one patients with intracranial lesions, planning to receive Cyberknife treatment, were enrolled. Each patient underwent CT with the same scanning parameters before and after administration of contract medium. For each patient 2 sets of DRR images (noncontrast, contrast) were generated as the references for the skull tracking image guided procedure. After image registering between DRR images and patients' digital images generated by image guided system of Cyberknife, paired t-test was applied on the results. Results: The deviations in 6 sets of position were all within ±0.02 mm. The max deviation between the two reference groups was 0.3 mm. All of the results have no statistical differences (P < 0.05). Conclusion: Using contrast medium or not for generating reference has little effect on accuracy of Cyberknife skull tracking. (authors)