RADIOLOGICAL OPTIMIZATION OF BUBBLE DYNAMICS IN CARBON DIOXIDE ANGIOGRAPHY

The using of carbon dioxide as contrast medium (CM) is increasing in the diagnosis of vascular diseases such as peripheral stenosis and fistulae. This radiological technique produces images through a digital subtraction (DS) of the images obtained with CM injected and a series of masks images. The two phase system (blood, CO2) remain separate by a surface until the gas is been dissolved. The radiological contrast of the xray image is given by integral dose on the pixels under the bubble respect to the pixel’s dose in the surround. Since the bubble is moving in the vessel, the fluoroscopy pulse time and the frame frequency change the contrast to noise ratio (CNR) of the contrast medium. This create moving artifacts that reduce the diagnostic outcome of the exam. To evaluate the best operational procedure we started with a laboratory model of the dynamical system. In this study has been realized a phantom made of 15cm of water with inside different glass vessels of various diameters. The images are taken with a PAXSCAN 4030 FP irradiated with RQR5 and RQR7 (IEC61267) x-ray field in different frame rate and integration time configuration. Moreover an algorithm to find the best radiological parameters has been developed to maximize the image quality indexes with no changing in entrance dose. The image quality improvement is evaluated in term of increment of the figure of merit of the image (CNR^2/Dose). The results shows the limits of the common used stacking technique and find out how long integration time protocols are less efficient than high frame rate protocols even in condition of high frame rate.