US 11,995,823 B2
Technique for quantifying a cardiac function from CMR images
Lucian Mihai Itu, Brasov (RO); Andrei Bogdan Gheorghita, Brasov (RO); Puneet Sharma, Princeton Junction, NJ (US); and Teodora Chitiboi, Jersey City, NJ (US)
Assigned to Siemens Healthineers AG, Forchheim (DE)
Filed by SIEMENS HEALTHINEERS AG, Forchheim (DE)
Filed on Aug. 17, 2021, as Appl. No. 17/445,204.
Claims priority of application No. 10 2020 211 736.5 (DE), filed on Sep. 18, 2020; and application No. 20465559 (EP), filed on Sep. 18, 2020.
Prior Publication US 2022/0092771 A1, Mar. 24, 2022
Int. Cl. G06T 7/00 (2017.01)
CPC G06T 7/0012 (2013.01) [G06T 2207/10088 (2013.01); G06T 2207/30048 (2013.01)] 21 Claims
OG exemplary drawing
 
1. A system for outputting a value indicative of an ejection fraction (EF), of a cardiac chamber of a heart based on a first temporal sequence of cardiac magnetic resonance (CMR) images of the cardiac chamber, the system comprising:
a computer configured as a neural network, the neural network comprising:
an input layer configured to receive the first temporal sequence of a first stack of slices of the CMR images along an axis of the heart, the first temporal sequence comprising one or multiple consecutive cardiac cycles of the heart; and
an output layer configured to output the value indicative of the EF of the cardiac chamber based on the first temporal sequence of the first stack of slices of the CMR images along the axis of the heart,
wherein the neural network has interconnections between the input layer and the output layer and was trained by a plurality of datasets, each of the datasets comprising an instance second temporal sequence of a second stack of slices of CMR images along the axis over one or multiple consecutive cardiac cycles for the input layer and an associated instance value indicative of the EF for the output layer.