Highly Accelerated Volumetric Brain Imaging with META and Deep Learning Reconstruction
Naoyuki Takei1, R Marc Lebel2, Suryanarayanan Kaushik3, Shohei Fujita4,5, Issei Fukunaga4, Shigeki Aoki4, Suchandrima Banerjee6, and Tetsuya Wakayama1
1GE Healthcare, Tokyo, Japan, 2GE Healthcare, Calagary, AB, Canada, 3GE Healthcare, Waukesha, WI, United States, 4Radiology, Juntendo University School of Medicine, Tokyo, Japan, 5Radiology, The University of Tokyo Graduate School of Medicine, Tokyo, Japan, 6GE Healthcare, Menlo Park, CA, United States
We combine advanced acquisition and reconstruction techniques for highly accelerated 3D brain imaging. A hybrid acquisition called META (Mixed Echo Train Acquisition) generates 3D T1W, T2W and FLAIR contrasts simultaneously and with high SNR efficiency. A 3D deep learning reconstruction (AIR Recon DL) reduces image noise, reduces artifacts, and enhances resolution relative to conventional reconstructions, solving the trade-off between SNR, scan time, and resolution. META with the DL Recon achieved 1mm isotropic T1W, T2W and FLAIR images within 4 minutes. These combinations are well suited in high resolution volumetric brain imaging.
This abstract and the presentation materials are available to members only;
a login is required.