High magnetic fields have potential advantages for SWI, due to higher contrast in both magnitude and phase images for tissues that contain iron-carrying paramagnetic substance. SWI is becoming more and more important as a clinical tool, since the resulting image gives excellent contrast between deoxygenated blood in veins and surrounding tissues. It can also give extra diagnostic information about pathologies concerned with many different neurological diseases. For example, micro-bleeds, iron accumulations, vascular malformations and calcifications [1]. In this study, we measure and compare contrast enhancement of top, middle, and bottom veins (e.g. cortical vein, anterior septal vein, hippocampal vein [2]) to investigate regional contrast enhancement differences for each number of phase mask multiplications (m) at 7T.All experiments in this study were done on the 7T MR system (Achieva, Philips Medical System, the Netherlands) with a single-channel quadrature transmit and 32-channel receive array coil (Nova Medical, MA, USA). This study included 5 healthy control subjects (mean±SD, 24.4±1.67 years). SWI data were obtained using gradient echo sequence. SENSE was used to accelerate the acquisition by an acceleration factor of 2. The following imaging parameters were used: TR=65 ms, TE=6ms, TA(α)=18, BW=176.7Hz/pix. FOV=220x220x140 mm³. Transverse orientation with a resolution of (0.5x0.5x2 mm³) was used (TA=6.3 min, 70 slices). After acquisition, DICOM magnitude and phase images were exported. The Philips scanner generated magnitude and filtered phase images. For SWI negative phase mask and sigmoid-shaped phase mask were applied [3]. To measure contrast in the images a line was drawn across the vein of interest (top, middle, and bottom of brain) [4]. Contrast enhancement was estimated [4] and an independent sample t-test was used to compare the degree of contrast enhancement between three veins of interest.In Figure 1, Top (a), middle (b), and bottom (b) brain slices of one healthy volunteer with selected vessels and 5-slice minIP of SWI are shown. Figure 2 shows intensity profiles of three vessels of all five subjects. As shown, profiles in top slice were uniform than those of middle and bottom veins because of not uniform susceptibility of others. In Figure 3, contrast enhancement using negative and sigmoid-shaped phase mask are shown as a function of number of phase masking applied to the original magnitude images. There was no significant contrast enhancement difference between middle and bottom over all m. However, there were significant differences between top and bottom (negative, m=1, p=0.0467; sigmoid, m=9 and 10, p=0.0416 and 0.0344), between top and middle (sigmoid, m=10, p=0.0410).We implemented SWI and contrast enhancement of three different veins were estimated and compared at 7T. Maximum mean contrast enhancement was 50/53% (negative/sigmoid, m=10/2) for the cortical veins (top), 32/27% (m=10/1) for the anterior septal veins (middle), and 26/24% (m=10/1) for the hippocampal veins (bottom). At 3T, CNR enhancement of cortical veins were reported as approximately 35% (m=4, TE=20 ms, 0.9x0.9x1.5 mm³ resolution) [3]. But not tried deep brain regions. Limitations of this study were that the first, image noise could not be measured so absolute CNR could not be estimated, but, in case of CNR enhancement, image noise cancel out so comparable with contrast enhancement. The second, intensity profiles were drawn manually. As shown in Figure 2, in the middle and bottom veins, getting uniform intensity profiles was difficult. To compare regional CNR enhancement difference, more robust intensity profiling, MR parameters optimization, and accurate image noise measurement would be needed.