Blood flow quantification by perfusion imaging or fractional flow reserve estimation using 2D x-ray angiography data requires scatter correction and beam hardening correction to ensure a linear relationship between the concentration of contrast agent in the human body and the x-ray image signal intensity. These types of correction usually require complicated calibration and numerous correction tables. To simplify correction, we developed a set of spherical epoxy resin phantoms containing different concentrations of barium sulfate. Prior to use, the concentration of iodinated contrast agent that results in the same x-ray absorption as each spherical phantom was determined. These values are hereinafter referred to as the equivalent concentrations of the spherical phantoms. To confirm that the spherical phantoms were suitable for estimating the contrast agent concentration, the curve of the image signal intensity vs. equivalent concentration was compared with the corresponding curve for the contrast agent. The mean difference between the two curves was 18 mgI/mL for 6 sets of experimental conditions. The mean of the p-values obtained in the paired t-tests was greater than 0.05. To validate the accuracy of our method, the spherical phantoms and test samples were placed on the surface of a chest phantom and an x-ray image including all the spherical phantoms and test samples was acquired. Using the relationship between the equivalent concentrations of the spherical phantoms and the image signal intensities, the concentrations of the test samples were estimated from their measured image signal intensities. The estimated and true concentrations of the test samples were then compared. Their cross correlation was R>0.99 for each of the 8 sets of experimental conditions used. The mean of the p-values obtained in the paired t-tests was greater than 0.05. Based on the promising results of these initial tests, we plan to proceed to clinical studies on these phantoms.