A Deep Learning Approach to Improve 7T MRI Anatomical Image Quality Deterioration Due mainly to B1+ Inhomogeneity
Thai Akasaka1, Koji Fujimoto2, Yasutaka Fushimi3, Dinh Ha Duy Thuy1, Atsushi Shima1, Nobukatsu Sawamoto4, Yuji Nakamoto3, Tadashi Isa1, and Tomohisa Okada1
1Human Brain Research Center, Kyoto University, Kyoto, Japan, 2Department of Real World Data Research and Development, Kyoto University, Kyoto, Japan, 3Diagnostic Imaging and Nuclear Medicine, Kyoto University, Kyoto, Japan, 4Department of Human Health Sciences, Kyoto University, Kyoto, Japan
The effect of transmit field (B1+) inhomogeneity at 7T remains even after correction using a B1+-map. By a deep learning approach using pix2pix, a neural network was trained to generate 3T-like anatomical images from 7T MP2RAGE images (dataset 1: T1WI and T1-map after B1+ correction and dataset 2: Inversion time [INV]1, INV2 and B1+-map). When the HCP anatomical pipeline was applied and compared, low regressions of original 7T data to 3T data were largely improved by using generated images by pix2pix, especially for dataset 2.
This abstract and the presentation materials are available to members only;
a login is required.