Steve Huntz Fung MD, Presenter: Nothing to Disclose

Yu-Bing Chang, Abstract Co-Author: Nothing to Disclose

Ramiro Pino, Abstract Co-Author: Nothing to Disclose

Edward Hui, Abstract Co-Author: Nothing to Disclose

Xiaobo Zhou, Abstract Co-Author: Nothing to Disclose

Miguel Valdivia y Alvarado, Abstract Co-Author: Nothing to Disclose

Goetz Benndorf MD, Abstract Co-Author: Nothing to Disclose

Orlando Diaz, Abstract Co-Author: Nothing to Disclose

Richard P. Klucznik MD, Abstract Co-Author: Nothing to Disclose

Zixiang Xiong, Abstract Co-Author: Nothing to Disclose

Stephen T. C. Wong PhD, Abstract Co-Author: Nothing to Disclose

King C. Li MD, Abstract Co-Author: Nothing to Disclose

To study feasibility of dynamic collimation during C-arm flat panel detector CT (FPCT) 3D rotational angiography (3DRA) to reduce radiation dose to patient, and to assess image quality of reconstructed collimated region-of-interest (ROI) by applying a cone-beam CT reconstruction algorithm with truncation correction.

Whole head 3DRA from a patient with left paraclinoid internal carotid artery aneurysm was acquired. Dynamically-collimated 3DRA was simulated by applying transverse collimation to each frame around a 7-cm cylindrical ROI centered on the aneurysm. Radiation dose to eyes, bone, brain, pituitary, and skin was calculated per frame without and with collimation using Monte Carlo simulation based on geometry and tube setting information from the C-arm system. Radiation dose was also measured in the center and periphery of a Perspex CTDI phantom using similar FPCT settings without and with static lead collimators to frame a 7-cm cylindrical ROI in the center of the phantom. Cone-beam CT reconstruction was performed using Feldkamp-Davis-Kress (FDK) algorithm without and with truncation correction based on raised-cosine windowed symmetric mirroring extrapolation.

Radiation doses calculated using Monte Carlo simulation were similar to those measured with the CTDI phantom. Simulation calculations indicate 70%, 60%, 45%, 38%, 27%, and 53% dose reduction to the contralateral eye, ipsilateral eye, bone, brain, pituitary, and skin, respectively, by applying dynamic transverse collimation. Phantom measurements also showed up to 60-70% dose reduction in the periphery with collimation. CT reconstruction of collimated ROI without truncation correction produced significant truncation artifacts. We found consistent near-complete removal of artifact after applying truncation correction with nearly identical gray level measured using voxel-by-voxel comparison between non-collimated and collimated CT images.

Dynamically-collimated whole head FPCT with truncation correction applied to CT reconstruction algorithm can reduce radiation dose by as much as 60-70% to various structures of the head, including eyes, while producing diagnostically-acceptable images similar to those without collimation.

Dynamically-collimated whole head FPCT with truncation correction can reduce radiation dose while producing diagnostically-acceptable images similar to those wtihout collimation.

Fung, S, Chang, Y, Pino, R, Hui, E, Zhou, X, Valdivia y Alvarado, M, Benndorf, G, Diaz, O, Klucznik, R, Xiong, Z, Wong, S, Li, K, Dynamically Collimated C-arm Flat Panel Detector CT Imaging for Focused Region-of-Interest Reconstruction and Radiation Dose Reduction.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11014044.html