Abstract
PURPOSE To determine whether signal variations and subtle anatomic deformities observed in high-resolution MR studies of temporal bones were caused by the large susceptibility differences at air-fluid interfaces near the round and oval window.
METHODS A systematic study of healthy subjects and plastic phantoms was conducted. The phantom consisted of a series of cylindrical holes of various small sizes within a solid block of plastic. These holes were partially filled with water and then covered with a reservoir of gelatin to simulate the otic capsule air-water interfaces. On a 1.5-T system, T2-weighted fast spin-echo images and three-dimensional Fourier transform gradient acquisition in steady state images were obtained using dedicated phased-array radio frequency coils. The directions of the frequency and in-plane phase-encoding gradients were swapped, and the receiver bandwidth was changed to demonstrate the dependence of the artifacts on these parameters.
RESULTS The phantom images confirmed and characterized artifacts consistent with magnetic susceptibility differences at the air-water interfaces. There is a combination of signal loss, misregistration in the frequency-encoding direction, and high signal foci related to the air-water interfaces. Furthermore, the artifacts were worse with narrower receiver bandwidth. Similar consistent artifact patterns were seen near the oval and round windows in studies of healthy subjects.
CONCLUSIONS In high-resolution MR imaging there are significant deformities in the display of the normal anatomy because of magnetic susceptibility.
- Copyright © American Society of Neuroradiology
