Purchase this article with an account.
Gilles de Hollander, Wietske van der Zwaag, Chencan Qiang, Peng Zhang, Tomas Knapen; Multi-center mapping of human ocular dominance columns with BOLD fMRI. Journal of Vision 2019;19(10):64b. doi: https://doi.org/10.1167/19.10.64b.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
The two eyes’ input into primate V1 is topographically organized into ocular dominance columns (ODCs). A few fMRI studies have shown the existence of ocular dominance maps in human V1, but only in a small subset of subjects with a flat V1 and using highly anisotropic ‘pencil’ voxels (Cheng et al., 2001 and Yacoub et al., 2007). Here, we report robust detection of ocular dominance columns using isotropic submillimeter voxels in the same subject, at two different 7 Tesla MRI sites separated by 7,815 km. We used gradient echo EPI (0.7mm 3dEPI and 0.8mm 2dEPI) on a Philips Achieva 7T scanner at the Spinoza Centre in Amsterdam and a Siemens 7T Magnetom at the Institute of Biophysics of the Chinese Academy of Sciences in Beijing. Using custom-built projection screens inside the transmit coil and prism glasses for dichoptic stimulus presentation, the subject was presented monocular counterphase flickering checkerboards for 8–24 seconds, in alternation between the left and right eyes. Using a highly automated segmentation, registration and preprocessing pipeline, we aligned the functional data to anatomical volumes and estimated percent signal change for left and right eye stimulation. In the Amsterdam dataset, the correlation between the left>right-contrast for the first 5 runs and the last 5 runs was r=.45. Between the datasets from Amsterdam and Beijing, there was a correlation between “left > right” contrasts r=.37 for all ‘signal’ voxels thresholded at p< 0.01 (uncorrected) in V1 in the Amsterdam dataset. These results show that ODCs from the same human subject can be reliably detected from two remote 7T machines with very different setups, using BOLD fMRI at submillimeter isotropic resolution. These approaches pave the way for future study of fine-grained functional architecture of the human brain.
This PDF is available to Subscribers Only