Abstracts

Previous MRI reports have suggested that small lesions of focal cortical dysplasia (FCD) may be located at the bottom of a sulcus. However, the exact relationship between FCDs and brain sulci remains unclear, mainly because of the complexity of the brain[apos]s convolution makes the visualization of gyral patterns difficult. Our purpose was to study systematically the spatial correspondence between small FCD lesions and brain sulci., We studied 40 consecutive patients with FCD. MRI was acquired on a 1.5T scanner using a T1-FFE sequence (1 mm³ isotropic voxels). Lesions were manually segmented by expert observers on T1-weighted MRI to calculate their volume (volume range: 128 mm³ to 94,620 mm³). Brain sulci were identified using an automatic extraction tool that includes skeletonization and automatic labeling of sulci. Segmented lesions and sulci were visualized both in 3D and 2D simultaneously. Lesions were considered to be located at the bottom of a given sulcus if spreading along both sulcal walls and centered on the pial surface of highest-curvature. Visibility on conventional MRI and lesion volume were used to divide FCD into small and large lesions. Volumes were thresholded using maximization of Gini coefficient, a disparity index., 16/40 (40%) patients had small FCD lesions. In 13 of them, lesions had been overlooked on MRI and had a volume ranging from 128 mm³ to 3093 mm³. 13/16 (81 %) FCD lesions were located at the bottom of a sulcus. Two others were related to the walls of several sulci, one was located at the crown of a gyrus. 15/16 lesions were located in the frontal lobe and 1 in the temporal lobe., Our results demonstrate for the first time a strong spatial relationship between small FCD lesions and sulci. Indeed, small FCD lesions are most likely located at the bottom of a sulcus. The inclusion of this information in automatic detection methods may improve the identification of subtle FCD and structural abnormalities in [ldquo]MRI-negative[rdquo] patients.[figure1], (Supported by the Canadian Institutes for Health Research (CIHR-203707).)