Juan Carlos Ramirez Giraldo MS, Presenter: Nothing to Disclose

Scott M Thompson BA, Abstract Co-Author: Nothing to Disclose

Bruce Knudsen MS, Abstract Co-Author: Nothing to Disclose

David Arthur Woodrum MD, PhD, Abstract Co-Author: Nothing to Disclose

Matthew Raymond Callstrom MD, PhD, Abstract Co-Author: Research Grant, Endocare, Inc Research Grant, Siemens AG

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

Physical phantoms that realistically mimic human vasculature and microvasculature are difficult, if not impossible, to design and build. The purpose of this study was to evaluate the feasibility of using a fixed ex vivo porcine liver to systematically evaluate dose reduction strategies in perfusion computed tomography (PCT).

A fresh porcine liver was perfused with formalin to preserve the specimen and ensure patency of the vascular structures. The portal vein was connected to a continuous flow pump, allowing control of input pressure and flow. The liver was submerged in a 32 cm wide, body-shaped acrylic phantom filled with water. The phantom was scanned four times, with a PCT protocol using 80 kVp and 400, 400, 100 and 20 mAs. 4 cc of iodine (300 mg/ml) was injected at 1 cc/sec for each scan and images were acquired for 50 sec, 2 sec after injection at each dose level. We compared percent root mean square (RMS) error between time attenuation curves (TACs) with respect to maximum CT attenuation minus background, and the accuracy of perfusion parameters such as blood flow (BF), blood volume (BV) and mean transit time (MTT) at three regions of interest (ROIs) to data measured with the first PCT scan (400 mAs).

The percent RMS error in TACs between the first 400 mAs scan and subsequent scans at 400, 100 and 20 mAs were 1.0%, 2.9% and 4.3% (input function) and 2.8%, 8.4%, 15.5% (tissue ROI), respectively. The average error across three tissue ROIs were 12.1%, 25.5%, 85.5% for BF, 6.45%, 7.1%, 68.2% for BV, and 10.5%, 36.9%, 31.4% for MTT, for the 400, 100 and 20 mAs acquisitions respectively.

The developed biological phantom provided highly reproducible TACs with percent RMS errors below 3%, and quantitative perfusion measurements of BF, BV and MTT agreement within 12% for identical PCT acquisitions. Low dose acquisitions revealed lower perfusion parameter accuracy.

The proposed phantom can be used to determine the lowest dose PCT scan protocols that provide accurate perfusion parameters.

Ramirez Giraldo, J, Thompson, S, Knudsen, B, Woodrum, D, Callstrom, M, McCollough, C, A Biological Liver Phantom to Evaluate the Impact of Radiation Dose Reduction on the Accuracy of CT Perfusion Imaging.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11008097.html