US 12,488,519 B2
Medical imaging conversion method and associated medical imaging 3D model personalization method
Benjamin Aubert, Québec (CA); Nasr Makni, Ferney Voltaire (FR); Thierry Cresson, Ottawa (CA); Carlos Alberto Vazquez Hidalgo Gato, Québec (CA); and Jacques A. De Guise, Québec (CA)
Assigned to EOS IMAGING, Paris (FR)
Appl. No. 17/924,630
Filed by EOS IMAGING, Paris (FR)
PCT Filed May 13, 2020, PCT No. PCT/IB2020/000508
§ 371(c)(1), (2) Date Nov. 10, 2022,
PCT Pub. No. WO2021/229253, PCT Pub. Date Nov. 18, 2021.
Prior Publication US 2023/0177748 A1, Jun. 8, 2023
Int. Cl. G06K 9/00 (2022.01); G06N 3/0464 (2023.01); G06T 7/00 (2017.01); G06T 7/10 (2017.01); G06T 11/00 (2006.01); G16H 30/40 (2018.01)
CPC G06T 11/008 (2013.01) [G06N 3/0464 (2023.01); G06T 7/0014 (2013.01); G06T 7/10 (2017.01); G16H 30/40 (2018.01); G06T 2207/10124 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/30012 (2013.01)] 28 Claims
OG exemplary drawing
 
3. A medical imaging conversion method comprising:
automatically converting:
at least one or more real x-ray images of a patient, including at least a first anatomical structure of said patient and a second anatomical structure of said patient,
both into at least a first digitally reconstructed radiograph (DRR) and a second digitally reconstructed radiograph (DRR) of said patient:
said first digitally reconstructed radiograph (DRR) representing said first anatomical structure without representing said second anatomical structure,
said second digitally reconstructed radiograph (DRR) representing said second anatomical structure without representing said first anatomical structure,
by a single operation using either one convolutional neural network (CNN) or a group of convolutional neural networks (CNN) which is preliminarily trained to, both or simultaneously:
differentiate said first anatomical structure from said second anatomical structure by isolating said anatomical structures directly in an original 3D volume, and
convert a real x-ray image into at least two digitally reconstructed radiographs (DRR) by one of:
(i) simultaneously producing a set of several DDRs, each of the DDRs focused on only one anatomical structure of interest, and
(ii) producing one DDR which is only focused on one anatomical structure of interest excluding the other anatomical structures.