Water fat separation using the single acquisition "sandwich" type 3-point Dixon method to optimize knee joint scans

In this paper, we tried to evaluate the effect of water-fat separation on and to optimize the scan condition of the newly developed "Sandwiched" 3-point Dixon method at 0.35 Tesla (T), for knee joint imaging. Using a 0.35T superconductive open magnet system with a solenoid type knee coil, one male and two female normal volunteers (27-37 y.o.) underwent knee joint imaging. Each sequence provided good water-fat separated images. At 0.35T, the gradient echo provided a better contrast than the spin echo. Optimal cartilage-marrow and cartilage-fluid contrast could be obtained at a frip angle (FA) of 90 degrees. There was no significant correlation between cartilage-marrow, cartilage-fluid contrast and repetition time (TR) values within the tested range. Cartilage-fluid and cartilage-marrow contrast were both best at an FA of 90 degrees with the gradient echo sequence. TR from 350 ms to 650 ms did not cause any significant contrast difference in the fat suppressed images. This method is useful and could be the only practical choice for obtaining fat suppressed T1 weighted images for joint magnetic resonance (MR) imaging at 0.35T.