A Study on CT Attenuation and MR Signal Intensity of Protein Solution

ABSTRACT

PURPOSE: To correlate CT attenuation and MR signal intensity with concentration of protein solution. MATERIALS AND METHODS: CT and MR examinations of a phantom containing bovine serum albumin solutions of various concentrations ranging from 0 to 55% were performed. CT Hounsfield units(HUs), MR signal intensities, and apparent diffusion coefficients (ADCs) of each albumin solution were measured, and CT HUs and MR signal intensities of the solutions were compared with those of cerebrospinal fluid (CSF), white matter, and cortical gray matter. RESULTS: CT HU increased gradually with increasing albumin concentration. On T1-weighted images(T1WI), signal intensity increased with increasing albumin concentrations of up to 35% but then decreased. On T2-weighted images(T2WI), gradually decreasing signal intensity and increasing albumin concentration were oibserved. Fluid-attenuated inversion recovery (FLAIR) and diffusion-weighted images (DWIs) showed that signal intensity peaked at a concentration of 10% and then gradually decreased. The ADC of the solution gradually decreased as concentration increased. Compared with those of normal brain structures, the CT HUs of solutions at concentrations of over 20% were higher than those of white and gray matter. At T1WI, the signal intensities of 10-45% solutions were similar to or higher than that of the gray matter. At T2WI, the signal intensities of solutions above 25, 35, and 40% were lower than those of CSF, gray matter, and white matter, respectively. FLAIR images showed that the signal intensities of 5-35% solutions were higher than that of gray matter. CONCLUSION: The CT attenuation of albumin solution increased gradually with increasing concentration. MR signal intensities peaked at 35% concentration on T1WI and at 10% on FLAIR and DW images, respectively, and then gradually decreased. T2WI and ADC map images showed gradually decreasing signal intensity and ADC as albumin concentration increased.