Four-dimensional multislice helical CT of the lung: qualitative comparison of retrospectively gated and static images in an ex-vivo system.

Abstract

To analyse the image quality of retrospectively gated helical CT using controlled respiratory motion of porcine lung explants. Five porcine lungs were examined inside a chest phantom. A silicone membrane was rhythmically inflated and deflated to simulate diaphragmatic respiration. Dynamic images (regular respiration at 8/min) and static scans (w/o respiration) at 0/25/50/75 and 100% of maximum inspiration were acquired with a 40-row detector CT scanner (rotation time 1s, pitch 0.1). Image quality on multi-planar reformations was evaluated by two observers. Partial projection artifacts, stepladder-artifacts and noise were compared for upper, middle and lower parts of the lung and different respiratory phases (scores 0-3 for absent, minimal, moderate and diagnostically relevant artifacts). Partial projection effects were limited to dynamic scans (mean score 1.33). Stepladder artifacts predominated in dynamic series compared to static series (mean score 0.55 versus 0.1; p<0.001). Image noise was not related to lung motion (mean scores 0.68-0.81). All artifacts predominated close to the diaphragm compared to the upper and middle parts of the lung (p<0.001 to p=0.02, respectively). Partial projection and stepladder artifacts were less in end-inspiration and end-expiration than within the respiration (p<0.001 and p=0.17, respectively). Diagnostically relevant artifacts were noted 9 times (9/9 close to diaphragm, 7/9 partial-projection). Even in ideal realistic conditions, helical 4D-CT produced tolerable artifacts which could be overcome by radiologists.