US 12,130,349 B2
Method for reconstructing a motion-corrected magnetic resonance image of a subject
Daniel Polak, Erlangen (DE); Stephen Farman Cauley, Somerville, MA (US); and Daniel Nicolas Splitthoff, Uttenreuth (DE)
Assigned to Siemens Healthineers AG, Forchheim (DE); and The General Hospital Corporation, Boston, MA (US)
Filed by Siemens Healthcare GmbH, Erlangen (DE); and The General Hospital Corporation, Boston, MA (US)
Filed on Nov. 21, 2022, as Appl. No. 17/990,895.
Claims priority of application No. 21210483 (EP), filed on Nov. 25, 2021.
Prior Publication US 2023/0160989 A1, May 25, 2023
Int. Cl. G01R 33/565 (2006.01); G01R 33/48 (2006.01); G01R 33/56 (2006.01)
CPC G01R 33/56509 (2013.01) [G01R 33/4818 (2013.01); G01R 33/5608 (2013.01)] 16 Claims
OG exemplary drawing
 
1. A method for reconstructing a motion-corrected magnetic resonance image of a subject, the method comprising:
providing k-space magnetic resonance data comprising a plurality of shots, wherein each shot of the plurality of shots corresponds to an individual motion state of the subject;
providing motion parameters related to each individual motion state of a plurality of motion states;
determining redundancies across the plurality of motion states of the plurality of shots based on the motion parameters, wherein the redundancies across the plurality of motion states are determined by finding similarities between different motion states, and wherein the finding of the similarities comprises comparing the motion parameters of the different motion states with each other and finding a redundancy when differences between the different motion states are below a predetermined threshold;
compressing the plurality of motion states based on the determined redundancies across the plurality of motion states; and
reconstructing a magnetic resonance image from the k-space magnetic resonance data based on the compressed plurality of motion states.