US 12,239,477 B2
Automatic determination of a motion parameter of the heart
Daniel Giese, Erlangen (DE); Jens Wetzl, Spardorf (DE); Maria Monzon, Bilbao (ES); Carola Fischer, Nuremberg (DE); and Seung Su Yoon, Erlangen (DE)
Assigned to Siemens Healthineers AG, Forchheim (DE)
Filed by Siemens Healthineers AG, Forchheim (DE)
Filed on Apr. 25, 2022, as Appl. No. 17/728,328.
Claims priority of application No. 21170677 (EP), filed on Apr. 27, 2021.
Prior Publication US 2022/0338829 A1, Oct. 27, 2022
Int. Cl. A61B 6/50 (2024.01); A61B 6/00 (2006.01); G01R 33/56 (2006.01); G01R 33/563 (2006.01); G06N 3/045 (2023.01); G06N 3/08 (2023.01)
CPC A61B 6/503 (2013.01) [A61B 6/5205 (2013.01); G01R 33/5608 (2013.01); G01R 33/56308 (2013.01); G06N 3/045 (2023.01); G06N 3/08 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A method for determining a motion parameter of a heart, comprising:
determining a sequence of cardiac magnetic resonance (MR) images showing a time resolved motion of the heart,
applying a subset of the sequence of cardiac MR images at a first input of a first trained convolutional neural network that is configured to determine, as a first output, a probability distribution of at least two anatomical landmarks in the subset,
cropping and aligning the sequence of cardiac MR images based on the at least two anatomical landmarks to determine a reframed and aligned sequence of further cardiac MR images,
wherein each one of the sequence of further cardiac MR images depicts the same orientation of the heart;
applying the sequence of further cardiac MR images at a second input of a second trained convolutional neural network that is configured to determine, as a second output, a further probability distribution of the at least two anatomical landmarks in each one of the sequence of further cardiac MR images; and
determining the motion parameter of the heart based on the second output.