Shuai Leng PhD, Presenter: Nothing to Disclose

Katrina Nesta Glazebrook MBChB, Abstract Co-Author: Nothing to Disclose

John Ignatius Lane MD, Abstract Co-Author: Nothing to Disclose

Kristin D. Zhao MA, Abstract Co-Author: Nothing to Disclose

Ryan Breighner, Abstract Co-Author: Nothing to Disclose

Thomas J. Vrieze RT, Abstract Co-Author: Nothing to Disclose

Cynthia H. McCollough PhD, Abstract Co-Author: Research Grant, Siemens AG

To assess noise, dose and spatial resolution of an ultra-high resolution (UHR) scan mode on a CT scanner equipped with an x-ray tube capable of small focal spot sizes at high mA, and to compare resolution to computed radiography (CR).

A human skull, a cadaver wrist, and a spatial resolution target were scanned on a 192-slice scanner (scanner A, Siemens Force) equipped with dynamic focal spot control, IR, and z-axis deconvolution, which allow a 0.4 mm image to be reconstructed from 0.6 mm detectors. All objects were scanned with 120 kV and 64x0.6mm collimation. The skull was scanned with 380 effective mAs (CTDIvol=45 mGy), the wrist with 200 effective mAs (CTDIvol=12 mGy), and the resolution target with 400 effective mAs (CTDIvol=24 mGy). For comparison, objects were scanned on a 128-slice scanner (scanner B, Siemens Flash) with 16x0.6 mm collimation and matched CTDIvol values. Images were reconstructed at the minimal thickness available (0.4 mm on scanner A and 0.5 mm on scanner B) with IR and high resolution kernels. For the skull and wrist images, sharpness was qualitatively evaluated by sub-specialty radiologists and image noise measured over uniform anatomic regions. CR images were acquired of the wrist and bar pattern, the limiting spatial resolution determined from the bar pattern, and all results compared between the two scanners and CR.

Images from scanner A were considerably sharper than those from scanner B. Two skull intralabyrinthine bone fragments were detected on scanner A that were not visible on scanner B. Trabecular bones of the wrist were better delineated on scanner A relative to scanner B, and were similar in appearance to CR. Image noise values on scanners A and B were 159 and 255 HU, respectively, for the head phantom, representing a 38% reduction in noise, and 71 and 73 HU for the wrist, respectively, representing a 8% reduction in noise. Limiting spatial resolution was 22 lp/cm for scanner A, 18 lp/cm for scanner B, and 22-25 lp/cm for CR.

Scanner A provided better spatial resolution and lower image noise compared to scanner B, and similar spatial resolution as CR. For the same CT image noise level, dose could be reduced on scanner A.

Spatial resolution comparable to CR can be achieved on CT systems, at noise or dose levels lower than previously available CT systems.

Leng, S, Glazebrook, K, Lane, J, Zhao, K, Breighner, R, Vrieze, T, McCollough, C, Demonstration of Dose or Noise Reduction, as well as Radiographic Spatial Resolution, on a Commercial CT Scanner.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14013420.html