Article Figures & Data
Figures
- Figure 1.
Preoperative axial structural magnetic resonance images of patient S.H. showing a tumor in the left temporal cortex. T1-weighted (A) and T2-weighted/FLAIR (B) axial sections are shown from top down and are aligned to the anterior–posterior commissure line. The z values indicate the distance (in millimeters) of a given axial section from the horizontal plane passing through the anterior and posterior commissures (z = −32 to 32 mm). Slices are shown as seen from above, i.e., the left hemisphere is shown on the left side.
- Figure 2.
Definitions and locations of the regions of interest (ROIs) constituting the hMT area and the hMST area of patient S.H. based on functional magnetic resonance imaging. BOLD activations were obtained by comparing responses to moving random dot kinematograms versus stationary dot patterns which were presented either in the left (A) or the right (B) visual hemifield. The resulting activations are superimposed on a structural axial slice in the plane of the ACPC, using a T-value threshold of 4.0 and a cluster threshold of k = 50 voxels (corresponding to p < 0.0001, uncorrected for multiple comparisons). A, Presentation of visual motion in the left visual hemifield induces a large patch of activation in contralateral (right) temporal cortex, corresponding to the hMT+, and a smaller patch in ipsilateral temporal cortex, defined as area hMST. B, Activations observed with the stimulus presented in the right visual hemifield. The pattern of activity is reversed. Note the mild posterior shift of area hMT+ on the left side due to the space-occupying tumor which is marked in green. C, Resultant ROIs attributed to area hMST (cyan) and area hMT (magenta), the first defined by responses to ipsilateral visual motion, the latter given by the voxels selectively responding to contralateral but not ipsilateral visual motion. D–I, ROIs of areas hMT and hMST superimposed on axial slices from top to bottom parallel to the ACPC plane (z = 8 to −12 mm). All axial images follow neurological convention, i.e., the left hemisphere is shown on the left. J, Three-dimensional surface reconstruction of the brain of patient S.H. and representation of the ROIs defining area hMT and area hMST. The ACPC plane (z = 0 mm) is indicated by blue lines.
- Figure 3.
Results of electrical stimulation on motion perception (A) and object naming (B). The surgical field giving view on the tumor (green) and temporo-occipital cortex before tumor resection including the electrical stimulation sites and the stimulation effects (left, anterior; right, posterior). A, Electrical stimulation sites visited in the visual motion direction discrimination task are given in magenta or cyan depending on the visual hemifield being stimulated (sketch above). Foci which induced motion blindness in the contralateral, i.e., right, visual field are indicated by solid magenta circles with black borders. Sites inducing motion blindness in the ipsilateral, i.e., left, visual field are indicated by solid cyan circles with black borders. B, Electrical stimulation sites in the object naming task. Foci inducing speech arrest are indicated by solid blue circles. The mean coordinates of the effective electrical stimulation sites for the three tasks are marked by crosses in the corresponding color. Solid white circles denote the stimulation sites at which no behavioral consequences for the given task was observed (contralateral visual motion, magenta; ipsilateral visual motion, cyan; object naming, blue). Anterior is on the left and ventral is at the bottom of the surgical field. An orthogonal reference grid with a resolution of 5 × 5 mm is indicated by blue lines.
- Figure 4.
Coregistration of radiological and intraoperative images was achieved by identifying common points of the trepanation defect in both modalities. A, Cropped and properly oriented three-dimensional surface reconstruction of the brain (dark gray) which was obtained from the preoperative MR images and which is overlaid here with the reconstructed skull (light gray) and its trepanation defect, the latter derived from postoperative CT images coregistered with the MR data. B, Picture of the surgical field giving view on the pial vessels and the cortical surface. In total 19 image tie-points were selected manually to delineate the trepanation defect (marked with cyan crosses in A and B). Anterior is on the left and ventral is at the bottom in both images.
- Figure 5.
Comparison of responses obtained from fMRI and electrical intraoperative stimulation. Coregistration of radiological and intraoperative images was achieved by identifying common points of the trepanation defect in both modalities. The reference picture of the surgical field including the stimulation sites (Fig. 3) was transformed to optimally fit into the trepanation defect such as reconstructed on the basis of postoperative computed tomography imaging. Coregistration of postoperative CT scans with preoperative fMR imaging in turn allowed superimposing of the ROIs of areas hMT (magenta) and hMST (cyan) onto the transformed surgical field, same color conventions as in Figures 2 and 3. A, Electrical stimulation sites at which contralateral (magenta) or ipsilateral (cyan) motion blindness was induced. B, Nonfoci. A shift of the electrical foci as compared with the BOLD responses is likely to reflect intraoperative displacement of the cortex in the course of the surgery (e.g., patient head positioning, skull trepanation, cerebrospinal fluid loss). The blue lines indicate the transformed reference grid of Figure 3.
