Anatomical Identification of V5 in Humans at 7 T

Introduction: Cortical areas are known to differ in their cytoarchitecture, myeloarchitecture, cortical thickness, vascularity, and regional transmitter/receptor distribution. The goal of this study was the identification of the visual motion area V5, also known as MT [1, 2, 3] as this area shows a regionally increased myelination. The earliest MRI study to show myeloarchitectural differences in grey matter demonstrated that the densely myelinated stria of Gennari demarcating primary visual cortex (V1) was MR visible [4]. Similar studies were performed at field strengths of 3 T [5, 6] and 4.7 T [7]. V5 has also recently been identified using structural MRI at 1.5 T [8, 9]. Using ultra-high field MR scanners it is now possible to examine the human cortex in vivo at a resolution of a few hundred micrometers with an adequate signal-to-noise ratio (SNR) and within a reasonable scanning time. In the present study a Turbo-Spin Echo (TSE) sequence was used for visualization of the cortical layers in V5 at 7 T. During a TSE echo-train any imaging slice experiences many off-resonant pulses, which generates magnetization transfer contrast (MTC) [10,11]. Combined with the T2-weighting and the proton density contrast (by using a relatively short echo time) MTC further helps to map myelinated cortical layers. Method: All experiments were performed on a 7 T whole-body MR scanner (MAGNETOM 7T, Siemens Healthcare Sector, Erlangen, Germany). An 8 channel phased array head coil (RAPID Biomedical, Rimpar, Germany) was used. All in vivo studies were carried out in accordance with ethics approval from the local university, and informed consent was obtained before each study. 17 coronal images covering the region where V5 is expected to lie were acquired using a first TSE protocol (TSE I: TR = 8.3 s, TE = 27 ms, refocusing flip angle = 120°, 2 averages, isotropic voxel size (0.5 mm)3, slice gap 0.25 mm). To assure that the visualized layer structure was not a motion artifact, a second set of TSE images was acquired with different parameters (TSE II: TR = 8.71 s, refocusing flip angle = 180°, slice gap 0.0 mm, other parameters were equal). The slice gap of zero allowed for a complete coverage of V5. Finally, a functional MRI experiment was performed to confirm the location of V5. A moving star field paradigm (block design: 8 repetitions moving star field vs. 8 repetitions static star field, 112 repetitions in total) and an EPI sequence (TR = 3.0 s, TE = 25 ms, flip angle = 90°, isotropic voxel size (1.4 mm)3, GRAPPA reconstruction, acceleration factor = 4) were used to functionally identify V5. Results: In Fig. 1 a coronal image acquired with the first TSE sequence shows a layer structure in V5 (highlighted by arrows). Four sections of coronal images acquired with TSE sequence II covering the extent of the layer structure in anterior-posterior direction are shown in Fig. 2. The structure is clearly visible in both hemispheres. The uppermost image (A) has the same slice position as the image shown in Fig. 1. Fig. 3 shows the activation pattern acquired with the described fMRI experiment overlaid on the same TSE images as shown in Fig. 2 (A)-(D). The activation pattern nicely corresponds with the region of the layer structure in the TSE images. Similar results were obtained with other subjects (not shown here).