US 11,730,388 B2
MRI image reconstruction using machine learning
Gerhard Laub, San Mateo, CA (US); Peter Schmitt, Weisendorf (DE); David Grodzki, Erlangen (DE); Waqas Majeed, Ellicott City, MD (US); and Wuyi Zhao, Shenzhen (CN)
Assigned to SIEMENS HEALTHCARE GMBH, Erlangen (DE)
Filed by Siemens Healthcare GmbH, Erlangen (DE)
Filed on Jun. 10, 2020, as Appl. No. 16/897,571.
Claims priority of application No. 201910554371.7 (CN), filed on Jun. 25, 2019.
Prior Publication US 2020/0405175 A1, Dec. 31, 2020
Int. Cl. A61B 5/055 (2006.01); G01R 33/56 (2006.01); G01R 33/563 (2006.01)
CPC A61B 5/055 (2013.01) [G01R 33/5608 (2013.01); G01R 33/5635 (2013.01)] 32 Claims
OG exemplary drawing
 
1. A computer-implemented method for reconstructing a MRI image, comprising:
acquiring first MRI measurement data at a first magnetic field strength using an MRI system;
generating a first MRI image based on the first MRI measurement data, the first MRI image having first imaging characteristics based on the first magnetic field strength;
generating a second MRI image by applying trained functions to the first MRI image, the second MRI image having second imaging characteristics, the second imaging characteristics corresponding to a second magnetic field strength different from the first magnetic field strength, the trained functions having been trained using training MRI images and reference MRI images, the training MRI images being obtained at the first magnetic field strength, and the reference MRI images being obtained at the second magnetic field strength; and
providing the second MRI image.