[Study of skin markers for magnetic resonance imaging examinations].

In magnetic resonance imaging (MRI), skin markers are used as a landmark in order to make plans for examinations. However, there isn't a lot of research about the material and shape of skin markers. The skin marker's essential elements are safety, good cost performance, high signal intensity for T1 weighted image (T1WI) and T2 weighted image (T2WI), and durable. In order to get a high signal-to-noise ratio (SNR) of T1WI and T2WI, baby oil, salad oil and olive oil were chosen, because these materials were easy to obtain and safe for the skin. The SNR of baby oil was the best. Baby oil was injected into the infusion tube, and the tube was solvent welded and cut by a heat sealer. In order to make ring shaped skin markers, both ends of the tube were stuck with adhesive tape. Three different diameters of markers were made (3, 5, 10 cmψ). Ring shaped skin markers were put on to surround the examination area, therefore, the edge of the examination area could be seen at every cross section. Using baby oil in the ring shaped infusion tube is simple, easy, and a highly useful skin marker.

[Study of image quality at various positions of reference scan].

In magnetic resonance imaging (MRI), we can use sensitivity encoding (SENSE) as a parallel imaging reconstruct from aliasing based on the sensitivity map of each coil element by reference scan. We researched SENSE reconstruction depending on the position of the reference scan. 1.5 T MRI (Achiva, Philips), 32-channel SENSE cardiac coil and rectangular solid phantom (nickel chloride) were used. Seven slices (A to G: from head to foot) were set at the phantom, slice thickness was 1 cm, and each slice interval was 5 cm. When T2 weighted axial images (TR: 3000 ms, TE: 120 ms) were taken, the center of the reference scan (C-Ref) was changed A to G, SENSE factor was changed 1 to 5, and number of signals averaged (NSA) was changed 1 to 3 with fold over suppression. Signal-to-noise ratio (SNR) and coefficient of variation (CV) from 9 regions of interest (ROI) of axial images were calculated. The SNR of slices A and G were lower than the other slices. %CV of slices F with C-Ref D was 69.4% lower than C-Ref A. %CV of slices D to F were higher than the other slices when C-Ref was A. %CV had no relation with SNR. Therefore, a 30 cm area around the center of reference can obtain best SENSE reconstruction.