US 11,810,228 B2
Network determination of limited-angle reconstruction
Michal Cachovan, Erlangen (DE); and Alexander Hans Vija, Evanston, IL (US)
Assigned to Siemens Medical Solutions USA, Inc., Malvern, PA (US)
Appl. No. 17/754,481
Filed by Siemens Medical Solutions USA, Inc., Malvern, PA (US)
PCT Filed Dec. 3, 2019, PCT No. PCT/US2019/064121
§ 371(c)(1), (2) Date Apr. 4, 2022,
PCT Pub. No. WO2021/112821, PCT Pub. Date Jun. 10, 2021.
Prior Publication US 2023/0162412 A1, May 25, 2023
Int. Cl. G06T 11/00 (2006.01)
CPC G06T 11/008 (2013.01) [G06T 2207/20084 (2013.01); G06T 2211/424 (2013.01)] 14 Claims
OG exemplary drawing
 
1. A system comprising:
a storage device;
a processor to execute processor-executable process steps stored on the storage device to cause the system to:
determine a plurality of sets of two-dimensional projection images, where each set of two-dimensional projection images includes two-dimensional projection images acquired from a plurality of projection angles;
from each of the plurality of sets of two-dimensional projection images, reconstruct an associated three-dimensional image volume;
from each of the plurality of sets of two-dimensional projection images, determine an associated subset of two-dimensional projection images, wherein a number of projection angles from which a given subset of two-dimensional projection images were acquired is less than a number of projection angles from which a given set of two-dimensional projection images associated with the given subset were acquired; and
train an artificial neural network to generate an output three-dimensional image volume based on input two-dimensional projection images, the training based on the subsets of two-dimensional projection images and the reconstructed three-dimensional image volumes.