Cone-Beam CT From Complete Data Using Saddle Trajectories On a Mobile Robotic CBCT Scanner
J Albrecht1,2*, S Rit3, P Steininger4, F Ginzinger4, P Huber4, I Messner4, M Kraihamer4, H Schmitz1, S Corradini1, C Belka1,5, C Kurz1, M Riboldi2, G Landry1, (1) Department of Radiation Oncology, University Hospital, LMU Munich, Munich, Germany, DE, (2) Department of Medical Physics, Ludwig-Maximilians-Universitaet Muenchen (LMU Munich), Garching (Munich), Germany, DE, (3) Univ Lyon, INSA-Lyon, Universite Claude Bernard Lyon 1, UJM-Saint Etienne, CNRS, Inserm, CREATIS UMR 5220, U1206, F-69373, LYON, France, FR, (4) Research & Development, medPhoton GmbH, Salzburg, Austria, AT, (5) German Cancer Consortium (DKTK), Partner Site Munich, Munich, Germany, DE
Presentations
SU-E-201-2 (Sunday, 7/10/2022) 1:00 PM - 2:00 PM [Eastern Time (GMT-4)]
Room 201
Purpose: With a novel, mobile robotic cone-beam computed tomography [CBCT] scanner, a 2-sin source trajectory, commonly known as a saddle, was implemented. This allowed reduction of cone-beam artifacts and exact reconstruction in the region of the convex hull for non-truncated data.
Methods: The medPhoton mobile ImagingRing [IRm] features a digital flat-panel detector and an X-ray source collimated to a cone-beam [CB]. The source and detector are independently movable on a tiltable ring gantry. Tilt up to ±30° allows the implementation of saddle trajectories, defined as having two periods of a sinusoidal trace on an imaginary sphere of radius 74.25 cm enclosing the region of interest [ROI]. To demonstrate CB artifact reduction, we scanned the head section of the Alderson phantom. The tilt maximum amplitude was varied between 10° and 25° using an acquisition time of 60 s, an energy of 120 kV and a filtration of 0.5 mm Cu. The detector setting was selected to 1440 x 1440 px², 12 fps and a gain of 5. Images were reconstructed using the conjugate gradient method from the RTK library, and compared to circular trajectory acquisitions.
Results: Scans performed with saddle source trajectory showed considerable reduction of CB artifacts in the convex hull of the source trajectory, when compared to circular trajectory reconstructions. The size of this convex hull where exact reconstruction for non-truncated data is possible was found in SI direction to be 13.4 cm [10° tilt] up to 35.3 cm [25° tilt] at the imaging center.
Conclusion: Using the saddle trajectory, demonstration of CB artifact reduction in the region of the convex hull was achieved, enabling reasonably large regions of exact reconstruction for non-truncated data in CBCT without drawbacks in scan time.
Keywords
Cone-beam CT, Image Artifacts, Phantoms
Taxonomy
IM- Cone Beam CT: Development (New Technology and Techniques)
Contact Email