|| Checking for direct PDF access through Ovid
Advantages of using a polymer contrast agent (CA) are i.e. their ability to work as a target specific CA. For this it is essential to enable detection of a thin layer of CA. However, preliminary results have indicated difficulties in visualizing polymeric CA with present contrast algorithms. The aim of this study was therefore to evaluate a new algorithm that improves the visualization of polymeric CA.US images of a tissue-mimicking phantom were obtained using GE Vivid7 (12L) and Philips iE33 (S5-1), Fig. 1. Repeated (n=34) contrast to tissue ratios (CTR) were performed at various MI (0.1, 0.4, 0.8) using the contrast algorithms pulse inversion (PI) and power modulation (PM) at a concentration of 10e5 MB/ml. A subtraction algorithm (SA) was developed based on subtraction of a reference - and a contrast image. The two images were matched spatially by using block-matching with normalized cross-correlation and in time by using the ECG. The possibility to detect a thin layer of CA was evaluated using the SA. The phantom was imaged upside down after 6 hours without flow, where the air filled CA had risen and produced a thin layer at the posterior vessel wall. The CTRs were 1.05 ± 0.05 (MI 0.1), 2.31 ± 0.16 (MI 0.4) and 1.72 ± 0.11 (MI 0.8) with PI, 1.00 ± 0.12 (MI 0.1), 0.79 ± 0.09 (MI 0.4) and 1.16 ± 0.13 (MI 0.8) with PM and 38.60 ± 2.45 (MI 0.1), 82.63 ± 4.07 (MI 0.4) and 100.55 ± 13.84 (MI 0.8) with the SA. Fig. 1d and 1e show long-axis images of the phantom with a thin layer of CA close to the posterior wall, which has been color-coded using the SA in Fig. 1e.The SA showed great improvements in CTR compared to existing contrast algorithms. The use of the SA to identify a thin layer of polymeric CA was initially tested, but will be further evaluated in future studies.