Program Information
'Pseudo-4D' Localizer Radiographs: A Novel Imaging Technique for Evaluating the Motion of Implanted Fiducial Markers at Simulation
D Thomas*, W Campbell , J Faught , B Jones , K Goodman , M Miften , University of Colorado School of Medicine, Aurora, CO
Presentations
SU-E-601-8 (Sunday, July 30, 2017) 1:00 PM - 1:55 PM Room: 601
Purpose: Evaluating the effectiveness of motion reduction techniques such as abdominal compression at simulation is challenging; while 4D-CT offers an evaluation of motion, repeated 4D-CTs at simulation are undesirable due to increased imaging dose, and fluoroscopy is rarely available for use at simulation. In this study, we demonstrate a technique to evaluate the motion of implanted gold fiducial markers, using low-dose fast-acquisition of “pseudo-4D”-localizer radiographs (‘scout’, ‘surview’ or ‘topogram’), similar to fluoroscopy imaging. This will allow image-based feedback of motion reduction techniques at simulation.
Methods: Repeated AP and Lateral localizer radiographs were acquired in alternating directions, and sorted slice-by-slice using RPM respiratory data. Phantom data was acquired using a motion stage and a phantom with embedded fiducial markers, with applied waveforms of varying amplitude and breathing-rates to evaluate the effect of motion blurring. Fiducial markers were identified in each image using template-matching technique, and the range of motion calculated. The technique was validated using digital phantom data acquired from patient fluoroscopy images, and the number of scans required to adequately capture the entire range of motion was evaluated.
Results: Template matching of the fiducial markers on each localizer radiograph allows automatic position identification and motion estimation. Acquiring 10-repeated arbitrarily acquired localizer radiographs allows an average of the 85th percentile of the entire free-breathing motion range to be measured, with low imaging dose (<4mSv compared to >30mSv for a typical 4D-CT).
Conclusion: Repeated fast localizer radiographs allow a low-dose method of evaluating tumor motion. The accuracy of the motion estimation depends on the actual positions of the markers acquired during each scan. Prospective triggering of the acquisitions based on the RPM signal would improve the accuracy of motion estimation. Further work will be performed to evaluate the potential for this technique in patients without fiducial markers, for example in lung tumors.
Contact Email: