US 11,842,465 B2
Systems and methods for motion correction in medical imaging
Guotao Quan, Shanghai (CN); Yi Wang, Shanghai (CN); and Jiao Tian, Shanghai (CN)
Assigned to SHANGHAI UNITED IMAGING HEALTHCARE CO., LTD., Shanghai (CN)
Filed by SHANGHAI UNITED IMAGING HEALTHCARE CO., LTD., Shanghai (CN)
Filed on Nov. 21, 2020, as Appl. No. 17/100,820.
Claims priority of application No. 201911146934.5 (CN), filed on Nov. 21, 2019.
Prior Publication US 2021/0158492 A1, May 27, 2021
Int. Cl. G06T 7/20 (2017.01); G06T 5/00 (2006.01); G06T 5/50 (2006.01); G06T 7/00 (2017.01); A61B 5/055 (2006.01); A61B 6/00 (2006.01); A61B 8/08 (2006.01)
CPC G06T 5/005 (2013.01) [G06T 5/50 (2013.01); G06T 7/0014 (2013.01); G06T 7/20 (2013.01); A61B 5/055 (2013.01); A61B 6/5264 (2013.01); A61B 8/5276 (2013.01); G06T 2207/10016 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/30048 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A system for motion correction in medical imaging, comprising:
at least one storage device including a set of instructions; and
at least one processor configured to communicate with the at least one storage device, wherein when executing the set of instructions, the at least one processor is configured to direct the system to perform operations including:
obtaining at least two image sequences relating to a subject, wherein each of the at least two image sequences is reconstructed based on image data that is acquired by rotating a radiation source of a medical imaging device by multiple gantry angles of the medical imaging device during one of at least two time periods, the subject undergoes a physiological motion during the at least two time periods, and the at least two image sequences satisfy a condition relating to a motion amplitude; and
generating at least one corrected image sequence relating to the subject by inputting the at least two image sequences into a motion correction model for correcting an artifact caused by the physiological motion, wherein the motion correction model is a trained deep learning model, a count of the at least one corrected image sequence is less than a count of the at least two image sequences, and the generating the at least one corrected image sequence relating to the subject by inputting the at least two image sequences into the motion correction model includes:
determining at least two ranked image sequences by ranking, based on the at least two time periods, the at least two image sequences;
inputting, according to the ranking of the at least two image sequences, the at least two ranked image sequences into the motion correction model; and
outputting the at least one corrected image sequence by the motion correction model.